Structures and Reaction Mechanisms of Organocuprate Clusters in Organic Chemistry

Home , Gilman reagent, Organocopper compound, Organonickel

REVIEWS

Wherefore Art Thou Copper? Structures and Reaction Mechanisms of Organocuprate Clusters in Organic Chemistry

Eiichi Nakamura* and Seiji Mori

Organocopper reagents provide the principles. This review will summarize example of molecular recognition and most general synthetic tools in organic first the general structural features of supramolecular chemistry, which chemistry for nucleophilic delivery of organocopper compounds and the pre- chemists have long exploited without hard carbanions to electrophilic car- vious mechanistic arguments, and then knowing it. Reasoning about the bon centers. A number of structural describe the most recent mechanistic uniqueness of the copper atom among and mechanistic studies have been pictures obtained through high-level neighboring metal elements in the reported and have led to a wide variety quantum mechanical calculations for periodic table will be presented. of mechanistic proposals, some of three typical organocuprate reactions, which might even be contradictory to carbocupration, conjugate addition, Keywords: catalysis ´ conjugate addi-

others. With the recent advent of and SN2 alkylation. The unified view tions ´ copper ´ density functional physical and theoretical methodolo- on the nucleophilic reactivities of met- calculations ´ supramolecular chemis- gies, the accumulated knowledge on al organocuprate clusters thus ob- try organocopper chemistry is being put tained has indicated that organocup- together into a few major mechanistic rate chemistry represents an intricate

1. Introduction 1 R Cu X

The desire to learn about the nature of elements has been R or R1 R and will remain a main concern of chemists. In this review, we R Cu will consider what properties of copper make organocopper R1 chemistry so useful in organic chemistry. O "RCu" R1 O Lying on the border between the transition metals and the [R2CuLi, RCu(X)Li, RMgX¥CuY, etc.] main group elements, copper occupies a unique position in the R R1 periodic table. The key roles of copper have been widely [1] recognized in various areas including superconductivity, O O R1X biological oxygenation,[2] and organic synthesis.[3±6] The most RÐR1 important utility of copper in organic chemistry is in the form of nucleophilic organocopper&i) reagents, which are used R either in a catalytic or a stoichiometric manner. Generally Scheme 1. Nucleophilic reactivities of organocopper reagents RCu. R sp2,sp3 carbon anionic centers; X, Y halogen, etc. formulated as R2CuM with a variety of metal and organic groups &M and R, respectively), metal organocuprates and related species are uniquely effective synthetic reagents for nucleophilic delivery of hard anionic nucleophiles such as reactions that can be achieved readily with organocuprate alkyl, vinyl, and aryl anions &Scheme 1). Conjugate addition,[7] reagents but not with other organometallics &Scheme 1). Half [11] carbocupration,[8] alkylation,[9] and allylation[10] represent the a century after Kharasch and Tawneys initial discovery, organocopper reagents are still the most useful synthetic [*]Prof. Dr. E. Nakamura, Dr. S. Mori reagents among transition metal organometallics.[12] The Department of Chemistry chemistry of organocopper reagents has been reviewed many The University of Tokyo times with emphasis on synthetic utility and sometimes Bunkyo-ku, Tokyo 113-0033 &Japan) [13] Fax : &81)3-5800-6889 on structural properties, but rarely on reaction mecha- E-mail: [email protected] nisms.[14]

Angew. Chem. Int. Ed. 2000, 39, 3750 ± 3771 WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000 1433-7851/00/3921-3751 $ 17.50+.50/0 3751 REVIEWS E. Nakamura and S. Mori

In spite of the long history and popularity of copper, there 1.1. Historical Background and Recent Progress remains the fundamental question: what properties of copper make organocopper reagents so useful? In a similar manner The initial implication of the forthcoming golden age of to Juliet on the balcony, we may say to ourselves ªO Copper, organocopper chemistry was in 1941, when Kharasch and O Copper, wherefore art thou Copper?º Information on the Tawney[11] reported the 1,4-addition reaction of a Grignard nature of reactive species in solution and their reactivities has reagent to an a,b-unsaturated ketone in the presence of a been fragmentary and incomplete. The most widely accepted small amount of a CuI salt.[24] Gilman et al. reported in 1952 ªresting stateº of the metal organocuprate&i) species in that addition of one equivalent of MeLi to a CuI salt results in solution is represented by the eight-centered dimer &R2CuLi)2 the formation of yellow precipitates, which then afford a shown in Equation &1), but there has been little consensus on colorless solution upon addition of another equivalent of MeLi &Scheme 2).[25] In 1966 Costa et al. isolated a complex between phenylcopper&i) and magnesium, as well as crystals of R Cu R ? ? a lithium diphenylcuprate&i) complex.[26] Although the or- Ð + Li ? Li + E+ RCuR /E RÐE + RCu (1) ganocopper reagents derived from Grignard reagents are ? RCuR widely used and may be described as R2CuMgX, it is still uncertain to what extent this reflects the reality in solution. the ªreactive conformation of a true reactive speciesº. To make matters worse, the structures of the final copper- containing products are generally unknown. Those exploring RLi RLi the frontiers of organocopper chemistry in industry and CuI RCu + LiI R2CuLi + LiI academia have desperately required a better mechanistic organocopper reagent organocuprate reagent understanding. Two sources of mechanistic information have surfaced in 2 RMgY CuX R2CuMgY¥MgXY the past several years, new analytical and theoretical methods. Scheme 2. Preparation of organocopper reagents. As recently demonstrated by Nakamura, Morokuma, and co- workers, theoretical analysis of the reactions of large lithium organocuprate clusters with realistic electrophiles has become The organic chemistry of organocuprates started its rapid possible with explicit consideration of solvent molecules and development in 1966, when House et al. showed that the bulk solvent polarity.[15±23] In this review, the current status of reactive species of conjugate addition is the lithium dior- understanding of organocuprate mechanisms will be summar- ganocuprate&i) called the ªGilman reagentº.[27] Foundations ized by focusing on lithium dialkylcuprate&i) clusters in the for subsequent vigorous synthetic development were laid by context of the following topics: &1) structures of the clusters in Corey and Posner,[28] and other important initial develop- crystals, in solution, and as studied by theoretical methods, ments, such as substitution reactions on sp2 carbons or in an &2) conventional mechanistic schemes for organocuprate re- allylic system[10, 29±31] and the carbocupration of acetylene[32] actions, and &3) reaction pathways of large lithium dialkylcu- were reported before the mid 1970s. prate&i) clusters as analyzed through a combination of The nature of the Gilman reagent now needs some careful theoretical and experimental data. We will suggest an answer definition. While numerous reports &in particular, old ones) to part of the question ªWherefore art thou copper?º describe the Gilman reagent as R2CuLi, a vast majority of

Eiichi Nakamura received his PhD in Chemistry from Tokyo Institute of Technology in 1978. Following postdoctoral study in the Department of Chemistry at Columbia University, he joined the faculty of his alma mater in 1980 and was promoted to the rank of full professor in 1993. In 1995, he moved to the Department of Chemistry of the University of Tokyo. His interest in the dynamism of action of organic molecules forms the basis for his research activities which range through organic synthesis, chemical theory, and biology.

Seiji Mori was born in 1968 in Tokyo. He studied at the Tokyo E. Nakamura S. Mori Institute of Technology, and was awarded his BSc in 1993 and MSc in 1995. In 1998, he recieved his PhD degree at the University of Tokyo under the guidance of Professor Eiichi Nakamura. He became a fellow of Japan Society for the Promotion of Science in 1997, worked at Emory University in USA, and is currently an assistant professor at Ibaraki University in Japan. His research interests center on the theoretical studies of organic and organometallic reactions.

3752 Angew. Chem. Int. Ed. 2000, 39, 3750 ± 3771 Organocopper Chemistry REVIEWS

them actually used an LiX cluster R2CuLi ´ LiX, prepared by Bu OP(O)(OEt)2 an in situ reaction between RLi and CuX &X Br, I, or CN, without Me3SiCl sometimes with a ligand such as Me S and PR ). Although 2 3 1.0 equiv BuTi(O-iPr)4Li R CuLi and R CuLi ´ LiX may show largely the same reac- + 5 mol% CuI¥2LiCl 2 2 O (2) tivities, Lipshutz et al.[33] showed by analyses of reactivities OSiMe3 and spectroscopic properties that they are indeed different Me3SiCl species. Even a small difference of solvent may affect the composition of the reagent and hence the reactivities. Due to Bu such complexity, it is now customary to show all ingredients [64] upon describing a reagent &for example, R CuLi ´ LiI ´ Me S/ catalyst, and a chiral ligand [Eq. &3)]. Magnesium-based 2 2 reagents have found use in quantitative fivefold arylation of BF3 ´Et2O in THF/hexane). Shortly after the opening of the ªLewis acid ageº with the discovery of the Mukaiyama aldol reaction in 1973,[34] O Et2Zn O Cu(OTf)2 (2 %) Yamamoto and Marayamas reports on the ªRCu ´ BF3º (3) reagent[35] introduced the new concept of Lewis acid assis- ligand (4 %) [36] tance in organocopper chemistry. Although the identity of C7H8, 3 h, -30¡C [37] 94 % RCu ´ BF3 is still elusive, the BF3 activation was applied to >98%ee numerous synthetic works, as illustrated in Scheme 3 by a diastereoselective addition of a homoenolate species in a total synthesis.[38] Since the initial discovery by Nakamura and Kuwajima in O CH3 [39] ligand = PN 1984, Me3SiCl has become a standard reagent for accel- O CH3 eration of conjugate addition. This effect was reported first for the copper-catalyzed conjugate addition of the zinc homoenolate of propionic acid esters, as shown in Scheme 3, then [40, 41] for Grignard-based catalytic reagents and stoichiometric [65] [66] C60 [Eq. &4)]and threefold arylation of C 70 ; this paves the [42±46] lithium diorganocuprate&i)s. way to new classes of cyclopentadienyl and indenyl ligands The zinc homoenolate started the chemistry of metal with unusual chemical properties.[67] homoenolates.[47±49] The synthetic scope of such a ªnucleo- phile-bearing electrophileº was further exploited by Tamaru, R Yoshida,and co-workers first,[50] and then extensively by [51±53] H Knochel et al., and others. Dominated by lithium- and RMgX/CuBr¥Me2S R R magnesium-based systems until the mid 1980s, organocopper (4) chemistry now routinely utilizes much milder organometallic >95% nucleophiles, such as organozinc, -titanium,[54] -zirconi- R R um,[55, 56] and -aluminum reagents.[57] With the aid of proper R = Ar, Me activators, these mildly reactive reagents show selectivities unavailable with the conventional reagents, as illustrated in [58] Equation &2) for Me3SiCl-dependent chemoselectivity. 1.2. Controversial Mechanistic Issues Organocopper chemistry is still rapidly expanding its synthetic scope. The scope of carbocupration, previously With the lack of detailed mechanistic information, various limited to acetylenes, has been extended to olefins.[59±62] fundamental mechanistic issues have remained the subject of Enantioselective conjugate addition[63] has become truly controversy. They are outlined in this section and some will be useful through the use of dialkylzinc, a cationic copper discussed later in more detail.

Scheme 3. An Me3SiCl- and BF3-accelerated catalytic conjugate addition reaction and cortisone synthesis. HMPA hexamethyl phosphoramide.

Angew. Chem. Int. Ed. 2000, 39, 3750 ± 3771 3753 REVIEWS E. Nakamura and S. Mori

[76] 1.2.1. The Single Electron Transfer Controversy for conjugate addition. On the other hand, BF3 ´Et2O accelerates conjugate addition[35] and alkylation of epox- House proposed in the 1970s[68] that conjugate addition of ides.[77] In the allylation chemistry, zinc-,[49] titanium-,[54] and organocuprate starts with single-electron transfer &SET) from aluminum-[78] based organocopper reagents show much higher the reagent to the enone substrate, and this hypothesis S 2' selectivity than lithium organocuprate. The Lewis acidity became widely believed for some time. After many years of N of the countercation to the organocuprate is undoubtedly studies, however, the evidence obtained for SET has proven to important, but the mechanistic role still needs further studies be too weak to substantiate this once-favored mechanism. [Eq. &7)].[18, 23]

1.2.2. The 1,2-Addition Mechanism in Conjugate Addition rate acceleration O by MXn R O Conjugate addition and carbocupration were once suggest- + "RCu" + MXn (7) ed to take place through direct 1,2-addition of RÀCu across a CÀC multiple bond [Eq. &5)].[69] As detailed in Section 4, recent theoretical studies have shown that the two reactions do take place in the same manner but not through this 1.2.5. Me3SiCl Acceleration mechanism. Acceleration and change of selectivity of conjugate additions by a silylating agent are now well established RCu [46] RCu RCu [Eq. &8)]. Me3SiCl also affects the stereoselectivity of RCulikewise? (5) carbonyl 1,2-addition.[79] Considerable mechanistic discus- O O sions have occurred in the literature.[80] One argument

rate acceleration O R OSiMe3 1.2.3. The Dummy Ligand + "RCu" + Me3SiX and silylation (8) A synthetic problem associated with the use of homoor- À ganocuprates [R2Cu] is that the reagent can transfer only one assumes simple Lewis acid activation of the starting enone [81] of the two precious R ligands to the target electrophile &E , with Me3SiCl, which is supported neither by experiments for example, an a,b-unsaturated carbonyl compound) and one nor by theory.[82] To the contrary, the lithium cation in the R ligand is lost as an unreactive RCu species. The introduc- organocuprate reagent has been shown to coordinate to À [70] [80a] tion, in 1972, of mixed organocuprates [R&Y)Cu] , in which Me3SiCl &on the chlorine atom). Chloride coordination to the Y group acts as a nontransferable dummy ligand, provided copper was suggested by theory[80b] on the basis of seemingly the first general solution to this problem [Eq. &6)]. Typical erroneous experimental data.[83] Another theorem, combined with an inner-sphere electron-transfer hypothesis, assumes in situ trapping of an enolate-like intermediate by the silylating R O agent, which makes the process irreversible.[42] The positive O R(Y)CuLi + CuY (6) correlation between the silylation power of the reagent and the magnitude of rate acceleration[83, 84] suggests that the rate- Y O + RCu determining step of the reaction is the silylation step rather than the CÀC bond-forming step. Mechanistic data such as reaction rate, stereochemistry, and theoretical analysis are still dummy ligands include alkynyl,[71] cyano,[72] phenylsulfanyl,[73] lacking, however. dialkylamino, and phosphanyl groups.[73, 74] The selectivity of ligand transfer has been thought to arise during the process of ligand ± ligand coupling in an intermediate bearing three 1.2.6. The Higher Order Organocuprate Controversy ligands, R, Y, and E. A widely accepted hypothesis has been that a Y group forming a stronger CuÀY bond acts as a better Organocopper&i) species bearing three anionic groups, 2À [27] dummy ligand &because it resists transfer). While this [R3Cu] , are called ªhigher orderº organocuprates. To hypothesis has been successfully applied to the design of allow differentiation, conventional organocuprate&i) species À dummy ligands, recent theoretical studies by Nakamura and [R2Cu] may be called ªlower order organocupratesº. It has Yamanaka revealed an entirely different controlling factor in been the subject of controversy whether or not there exists, as the dummy ligand chemistry &see Section 4.2).[22] a stable species, a higher order lithium cyanocuprate

ªR2Cu&CN)Li2º that bears two carbanionic residues and a cyanide anion on the copper [Eq. &9)]. The controversy 1.2.4. Lewis Acid Effects of the Countercation generated numerous mechanistic and structural studies on It has been shown for all major categories of lithium organocuprates in general. organocuprate reactions[15, 17, 75] that addition of a crown ether significantly retards the desired reactions. In addition, sodium organocuprates are much inferior to lithium organocuprates

3754 Angew. Chem. Int. Ed. 2000, 39, 3750 ± 3771 Organocopper Chemistry REVIEWS

I It was reported in the 1970s that a higher order organo- and R3Cu Li2 generated in situ are the most useful sources of cuprate reagent, which was prepared by the use of more than carbon nucleophiles and hence have been studied most III two equivalents of an alkyllithium reagent for a copper&i) salt, extensively. The structure of the R3Cu species, which has is more reactive[27, 85, 86] and more selective than ordinary long been discussed in the literature without rigorous proof of organocuprates.[87] Ashby and Watkins showed by NMR its existence, was recently determined. A summary of the spectroscopy and cryoscopy that species form which may be structures of organocopper reagents will be given below. considered as higher order organocuprates.[88] Bertz and Discussions will be limited to those pertinent to the under- Dabbagh demonstrated for the first time the presence of a standing of organocuprate reaction mechanisms, as more I À [13] tricoordinated Cu complex &[R3Cu2] ) by solution NMR extensive structural reviews have already been published. spectroscopic studies,[89] and Olmstead and Power[90]demons- Several representative crystal structures are shown in Fig- trated the existence of a tricoordinated organocuprate ure 1, and will be discussed in the following sections. [91] [Li3Cu2Ph5&SMe2)4]in crystals. Lipshutz et al. reported that reagents formed by addition of two equivalents of RLi to CuCN are higher yielding than the 2.1. Mono- and Dicoordinated Organocopper0i) corresponding Gilman organocuprates &R CuLi) or lower 2 Compounds order cyanocuprates &RCu&CN)Li), and described it as R Cu&CN)Li to imply a tricoordinated structure.[72b, 92] With 2 2 2.1.1. Neutral Organocopper,i) Compounds: RCu the aid of 13C, 6Li, and 15N NMR data,[93, 94] Bertz et al. pointed out that cyanide is not attached to copper&i) in the Lipshutz Though MeCu itself is polymeric in ethereal solution, RCu [95] mix, and started the controversy. Physical measurements by may form a monomeric cluster such as [MeCu&PPh3)3]in the Penner-Hahn and co-workers[96, 97] and Lipshutz et al.,[98] and presence of a ligand having high affinity to copper &such as theoretical studies by the groups of Snyder,[99] and Penner- phosphane; Figure 1 a).[114] A strong neutral donor ligand &like [97] Hahn and Frenking contributed much to the discussion. All PPh3) may increase the reactivity of a neutral organocopper of the recent crystallographic data for the cyanocuprates of reagent.[115] Neutral organocopper&i) compounds have also ªhigher order stoichiometryº reported by Boche et al.[100] and been characterized as eight-centered cyclic tetramers, for [101] [116] [117] van Koten and co-workers indicated that the cyanide anion example, [&2-Me2NC6H4)Cu]4, [&Me3SiCH2)Cu]4, and [118] is coordinated to lithium and not to copper. Consensus after [PhCu]4 ´2Me2S &Figure 1b). We see here that a copper many years of studies is therefore that three-coordinated atom in a neutral RCu compound may accept a solvent 2À [R2Cu&CN)] is not a stable structure in ethereal solu- molecule as a ligand. It is a general characteristic of neutral tion.[94, 102±104] In spite of this conclusion, the Lipshutz mix still RCu aggregates that we find a three-center two-electron remains as the reagent of choice in many synthetic trans- bonding pattern and the copper atoms solvated with a formations,[105] and the presence of a tricoordinated cuprate&i) heteroatom ligand. dianion was indicated by 13C±13CN carbon coupling for the Since the method commonly used for generation of RCu cyanostannylvinylcuprate&i) dianion in a mixture of THF/ species produces an extra alkali metal salt in solution HMPA.[106] In addition, the cyanide anion finds its way to &Scheme 2), the reagent commonly described as a neutral copper at the end of the reaction and forms RCu&CN)Li, RCu reagent may react as a halocuprate complex. In fact, although it is not known when the cyanide ± copper coordi- density functional calculations indicate that conversion of free nation starts. The true role of the cyano group in the reactions monomeric MeCu into a six-centered cluster MeCu ´ 2LiCl is of higher order cyanocuprates still remains obscure.[103] an energetically favorable process [Eq. &10)].[17] In addition,

1.2.7. Other Issues Cl Cl Li Li Me Cu + Li Li (10) While a large number of studies have reported on conjugate Cl Me Cu Cl addition and SN2 alkylation reactions, mechanisms of many important organocopper reactions have not been discussed. the crystal structure of a mixed alkylhalocuprate bearing an These reactions involve substitution on sp2 carbons,[28] acyla- RÀCuÀX moiety was reported recently &Figure 1c).[119] The tion with acyl halide,[107] carbonyl addition, oxidative cou- particular compound reported was isolated as a monomeric pling,[108] nucleophilic opening of electrophilic cyclopro- species owing to the bulkiness of the organic group. panes,[109] and the Kocienski reaction.[110] The chemistry of organocopper&ii) species has rarely been studied experimentally[111±113] or theoretically. À 2.1.2. Linear Free Organocuprates: [R2Cu] The most basic structural property of diorganocuprate&i)is 2. Structures of Organocopper Compounds their linear RÀCuÀR arrangement. It is clearly seen in À [Li&[12]-crown-4)2] [CuPh2] ´ THF, prepared by mixing [120] Organocopper compounds can be classified into four basic Ph2CuLi and the crown ether &Figure 1d). A linear I I I III À[120] types, RCu ,R2Cu Li, R3Cu Li2 , and R3Cu . As represented organocuprate structure of [Me2Cu] was also identified by polymeric MeCu, the RCu reagents are unreactive and not in crystals. Note that these solvent-separated ion pairs are very useful synthetic reagents. R2CuM &M Li, MgX, etc.) unreactive in many standard organocopper reactions, the

Angew. Chem. Int. Ed. 2000, 39, 3750 ± 3771 3755 REVIEWS E. Nakamura and S. Mori

Figure 1. Crystal structures of representative organocopper compounds. Hydrogen atoms are omitted for clarity. In &g) and &h) the dihedral angles are given.

Trip 2,4,6-iPr3C6H2 .

3756 Angew. Chem. Int. Ed. 2000, 39, 3750 ± 3771 Organocopper Chemistry REVIEWS reasons for which have recently been elucidated by the studies While no crystal structures have been determined for [15] À of Nakamura, Morokuma and co-workers. R2CuLi ´ LiX &X halogen), the presence of &LinXn1) units in organocuprate clusters has been suggested through studies using electrospray ionization mass spectrometry.[133] The six-

2.1.3. Minimum Lithium Organocuprate Clusters: R2CuLi centered structure of a neutral arylcopper complex &Fig- ure 1i)[116] offers a good model, and is similar to the structure Coordination of a lithium cation to the linear RÀCuÀR of Me CuLi ´ LiCl or Me CuLi ´ LiI determined by calcula- anion creates RÀCuÀRÀLi, which may be called a minimum 2 2 tions. lithium diorganocuprate&i) cluster. For the partial charges on In the hypothetical structural conversion shown in the two ends to be neutralized, the minimum cluster Scheme 4, we may start with a free dialkylcuprate&i) anion dimerizes, polymerizes, or attracts an ambiphilic LiX unit to [R1R2Cu]À and go through a minimum cluster R1R2CuLi to form a closed cluster, R CuLi ´ LiX.[121] Due to the high energy 2 either polymers, eight-centered dimers, or mixed cyclic cost of bending the RÀCuÀR bond,[21] the four-centered cyclic clusters. The linear R1ÀCuÀR2 structure will be retained structure &often found for alkyllithium compounds) does not throughout the conversion. The stable linear CÀCuÀC form. geometry is due to bonding participation of the copper d Monomeric [Li&thf) ][CuMe&PtBu )]&Figure 1e) [122] has z2 3 2 orbital.[21] been observed in crystals obtained from a reagent mixture that shows the typical reactivity of an organocuprate bearing a dummy ligand. In the structure of linear dimers of lower order R1 Cu + δ− [123] + Li Li cyanocuprates ArCu&CN)Li &Figure 1 f), we also see 1 2 δ+ R Cu R Li strong LiÀN association. Such a dimeric unit structure may R1 Cu R2 2 [100] Cu R be found in a polymeric complex. Li R1 R2 + LiX Cu R2 Cu R1 X 2.1.4. Closed Clusters: R CuLi ´ LiX and ,R CuLi) Li 2 2 2 Li Li Li Li R1 Cu R2 X R1 Cu R2 1 2 The neutral closed cluster is widely detected in crystals and Li R Cu R in solution, and accepted as the typical resting state of lithium Scheme 4. Various structural possibilities of cuprates. Solid lines indicate organocuprates in ethereal solution. Eight-centered dimers of &largely) covalent bonds and dashed lines indicate &largely) electrostatic bonds between a metal cation and an organic or heteroatomic anion. &Such R2CuLi were suggested first by van Koten and Noltes on the basis of NMR spectroscopic and cryoscopic analyses of distinctions will not be generally made in the remaining schemes in this review.) X RCuR, halogen, CN, etc. lithium diarylcuprate&i) compounds in benzene.[124] The dimeric structure was further indicated in NMR spectroscopy, cryoscopic, and X-ray scattering experiments by Pearson and Gregory.[125] Following the first crystal structure determina- [126] 2.1.5. Dynamic Equilibria in Solution tion of a R2CuLi by van Koten et al., Weiss and Lorenzen, and Olmstead and Power determined various dimeric Although the crystal structures offer important reference [127] structures [{[Li&OEt2)]CuPh2}2]&Figure 1g), [Li2&SMe2)3- points, they do not reflect the reactive conformations of these [118] [128] Cu2Ph4], and [{[Li&SMe2)]Cu&Me3SiCH2)2}2]. These species in solution. Recent experimental studies revealed a structures have their lithium cation tricoordinated &two R wealth of information on the time-averaged, as well as groups and one solvent molecule). The R groups &aryl and dynamic, structures in solution. alkyl) are pentacoordinated with a stronger bond to copper The Me2CuLi dimer is the species studied most carefully. Its and a weaker electrostatic bond to lithium. As seen from the molecular weight indicates a &time-averaged) dimeric struc- metalÀCipsoÀCorthoÀCmeta angles in lithium diphenylcuprate ture in diethyl ether and Me2S, and X-ray scattering studies &Figure 1g), the copper atom is s-bonded and the lithium support this conclusion by showing the CuÀCu distance in atom is p-coordinated to the ipso phenyl carbon atom. In a solution as 2.6 ± 2.8 ,[125, 130] which is close to the value in dialkylcuprate&i), the copper atom is covalently bound to a crystals. The extended X-ray absorption fine structure &EX- nearly tetrahedral sp3 carbon, which is electrostatically AFS) spectra in solution indicate dicoordination of the copper associated with a lithium atom. Unlike the copper atom in atom with a CÀCu bond length of 1.94 ± 1.96 .[130, 134] All neutral RCu complexes &see Section 2.1.1.), the negatively these data are consistent with the persistence of the eight- charged copper atom is not Lewis acidic and does not accept centered structure both in crystals and in solution. [129, 130] an extra ligand &solvent). Cryoscopic measurement of the reagent Me2CuLi ´ LiI in A structure of an organomagnesium-based organocopper&i) THF indicated persistence of this stoichiometry,[135] whose compound is known and has a magnesium-bridged structure stable structure was shown by theory to be the six-centered &Figure 1h).[131] As seen from the dihedral angle in Figure 1h, structure shown in Scheme 4. Supporting such a structure, the Lewis acidic magnesium atom is more tightly bound to the EXAFS spectroscopic studies on Me2CuLi ´ LiCN and Me2- phenyl group than than the lithium in the lithium organo- CuLi ´ LiI in THF indicated a single Cu atom in a clus- cuprate. Organocuprates bearing more Lewis acidic metals ter.[96a, 130, 136] According to NMR spectroscopy studies in such as zinc&ii) and titanium&iv) cations have not yet been diethyl ether,[137] the cuprate may exist as an equilibrium [132] characterized. mixture of the Me2CuLi dimer and the LiX cluster.

Angew. Chem. Int. Ed. 2000, 39, 3750 ± 3771 3757 REVIEWS E. Nakamura and S. Mori

Recent NMR studies by Berger and co-workers on the Putting together the body of current data, one can draw a cuprates Me2CuLi, Me2CuLi ´ LiMe, and Me2CuLi ´ LiCN in series of mobile equilibria, as shown in Scheme 5, for

THF revealed the dynamic properties of the clusters in &Me2CuLi)2 and Me2CuLi ´ LiX; they will dissociate sequen- [104] À solution. The lithium atom exists in the time-averaged tially into free [Me2Cu] with Gibbs free energy loss in À vicinity of the methyl group in the [Me2Cu] moiety, but ethereal solution. Me2CuLi ´ LiX equilibrates with the dimer rapidly exchanges with other lithium atoms. Hence, a through exchange of an Me2CuLi unit with LiX. Oligomers predominant and thermodynamically stable cyclic cluster and polymers may also form. As will be discussed in Section 4, undergoes fast and reversible cleavage of the MeÀLi electro- the ªopen clusterº &among other clusters) serves as a gateway static bond &Scheme 5). The spectroscopic observations channel to organocuprate reactions. are consistent with the experimental and theoretical thermo- dynamic considerations. Dissociation of the &Me2CuLi)2 dimer to the monomer is enthalpically highly disfavored by I 2À gas-phase calculations,[138] and must also be entropically 2.2. Tricoordinated Copper Compounds: [R3Cu ] and III disfavored due to solvent participation,[139, 140] as is known R3Cu experimentally for alkyllithium oligomers &see Equation &11); the dissociation process is accompanied by a loss of 2.2.1. Higher Order Organocuprates À1 À1[140] 18.8 cal mol K ). [90, 91] [Li3Cu2Ph5&SMe2)4]&shown in Figure 1j) is the only crystal structure known for the controversial[89] nucleophilic higher order organocuprates. This complex forms only from a (BuLi) ¥4THF+ 4 THF 2[(BuLi) (thf) ] (11) 4 2 4 dimethylsulfide solution, and not from diethyl ether or THF. Therefore it is not clear if the higher order organocuprates can The presence of lithium halide is known to affect reactiv- also exist in ethereal solution. Solution NMR studies[104] [33] ities and selectivities, but only to an energy scale of a few indicated that Me3CuLi2 dissociates into Me2CuLi and MeLi kcalmolÀ1 of activation energy. This may be partly due to through rapid exchange of lithium atoms. As to the contro- accelerated cluster reorganization or to coordination of versial higher order cyanocuprates, no such tricoordinated lithium halide to a CuIII intermediate, which is formed during copper&i) species have been experimentally or theoretically 2À nucleophilic reactions. In the light of contrasting reports &one identified, except a case reported for [Cu&CN)3] &Fig- reports slower conjugate addition in THF than in diethyl ure 1k),[144] where the two negative charges are stabilized by ether,[141] another reports a faster reaction in toluene,[142] and the presence of three cyano groups. Other than this tricyano the further one reports that 1,4-addition in toluene can be structure, all known organocuprates of a higher order [143] promoted over 1,2-addition in the presence of Me2S ), cyanocuprate stoichiometry have usual dicoordinated copper solvent effects are a difficult subject to deal with. centers.[100, 101, 123]

closed cluster open cluster minimum cluster free cuprate

Me S Me Cu Me S Me Cu Me Cu Li +2S S +2S +2S S Li S Me S S S Li S S Li Li S SH X 2 S Li S S S Li Me Cu Me Me Cu Me Me Cu Me Me Cu Me +LiX S S S S X +2S +2S +S S Li S XSLi X Li S S XSLi SSLi Li S S S Li S S Li S S Me Cu Me Me Cu Me Me Cu Me Me Cu Me

S Li -2S polymers polymers X Me Cu Me

Scheme 5. Rapid cluster equilibration for &Me2CuLi)2 and Me2CuLi ´ LiX. Dashed lines indicate coordination between a metal cation and a neutral solvent molecule or ligand; S solvent; X halogen, CN, RO, etc.

3758 Angew. Chem. Int. Ed. 2000, 39, 3750 ± 3771 Organocopper Chemistry REVIEWS

2.2.2. Organocopper,iii) Species categories of organocuprate reactions, conjugate addition, carbocupration reactions, and alkylation reactions, have been Although trialkylcopper&iii) species have appeared in most extensively studied. numerous mechanistic discussions since the 1970s, it is only since the late 1980s that crystal structures substantiated the presence of such species. We now have three crystal structures 3.1. Conjugate Addition of discrete organocopper&iii) species.[145] As illustrated for [Cu&CF H) ]À in Figure 1l,[145c] all of them are nearly square 2 4 3.1.1. The Four- and Six-Centered Mechanisms planar in geometry in consonance with the formal d8 electronic system of the copper atom, and they can be Four-centered addition of RCu to an enone was discussed in regarded as T-shaped neutral CuIII species bearing a fourth the 1960s.[162] There have been discussions about a six- III centered transition state until recently,[152] but this does not ligand. Recent theoretical analysis on the stability of Me3Cu [146±149] supported the stabilizing effects of a donor ligand. explain however the formation of an E/Z mixture of enolate stereoisomers.[159, 160] These mechanisms &Scheme 6) must now be considered obsolete. 2.3. Theoretical Analysis of Organocopper Structures MX O RCu + R R O Although a variety of structural data were already accu- O RMX O OMX ? RCu mulated by the beginning of the 1990s, connecting this mass of Cu Cu Cu R structural data into molecular level mechanisms has been a R a) b) difficult task. With the advent of theory and computers, R quantum mechanical calculations are now playing the key Scheme 6. a) 1,2-Addition and b) 1,4-addition mechanisms. role. The development of post Hartree ± Fock &HF) and density functional calculations, which was honored by the 1998 Nobel Prize, has made a strong impact on the study of 3.1.2. The Single Electron Transfer Theorem organocopper chemistry. House and co-workers pioneered the synthetic and mech- Pioneering theoretical works by Poirier et al.,[150] Ziegler anistic studies of organocuprate reactions in the 1970s. In their et al.,[151] Morokuma et al.,[152, 153] Bauschlicher et al.,[154] papers a mechanism was proposed that assumes a single Frenking et al.,[155] and Baerends et al.[156] often used mono- electron transfer &SET) from the dimer which would lead meric MeCu. Studies by Hwang and Power, Snyder et al., to a CuIII intermediate &Scheme 7).[68, 161] The SET/CuIII theo- Sosa et al., and Frenking and co-workers demonstrated the importance of electron-correlated calculations in the structural studies of organocuprate clusters.[119, 121a, 157, 158] The O electron-correlated methods &MP2, B3LYP, and higher levels) R Li R are mandatory for description of the intermediates and Cu Cu R Li R O SET R Li R transition states in the reactions of lithium organocuprate I I Cu Cu R Li R clusters. Good description of the high-lying copper 3d orbitals R Li R CuI CuI O is necessary. The HF calculations give very low-lying dxz orbitals, and understandably fail in adequate description of R Li R p-backdonation ability of the cuprates.[19] As is the case for R Zn, neutral RCu has 3d orbitals that are too low-lying to be reductive 2 R Li R nucleophilic, and it behaves rather as a Lewis acid.[19] elimination Cu III R Relativistic effects on copper are small on structures and Cu R Li R energetics.[15] The high activation energies of MeCu addition O O reactions obtained in earlier studies[115, 156, 157] are due to lack Scheme 7. The House mechanism of 1,4-addition. of d-orbital participation. rem had a strong impact for many years. However, most of the experimental facts, once considered to support the SET 3. Experimental Studies on Organocuprate process, are not accepted as evidence of SET now. Only the Reaction Mechanisms CuIII hypothesis has survived the test of the time. 1) E/Z Isomerization of the olefinic part in an enone was Mechanistic studies on organocuprate reactions have been once taken as evidence for reversible electron transfer. It hampered by the complexity of the cluster structures. was later reported that the isomerization takes place even Crystallographic and spectroscopic structural analyses have in the presence of LiI, a usual component of the Gilman [163] provided information only for discrete data points. Various cluster reagent &that is, Me2CuLi ´ LiI). Such isomer- contradicting mechanistic proposals have made obscure the ization is also possible by reversible generation of an mechanistic pictures of organocuprate chemistry. Nonethe- advanced d !p* copper/enone complex along the reaction less, a certain mechanistic consensus has been achieved, and is pathway,[42, 164] and, thus, does not represent strong evi- summarized in the following sections. Three important dence of SET.

Angew. Chem. Int. Ed. 2000, 39, 3750 ± 3771 3759 REVIEWS E. Nakamura and S. Mori

2) Qualitative correlation of the apparent rate of 1,4-addition with the reduction potential of the enone[161] was later proven by quantitative kinetic studies by Krauss and Smith to be only superficial.[165] 3) b-Cyclopropyl a,b-unsaturated ketones, such as the one shown in Equation &12), often give a ring-opened product, which was taken as strong evidence for radical anion formation by SET.[166] An elegant study by Casey and Cesa This rate expression is consistent with the reaction scheme with deuterium-labeled substrate indicated stereospecific- shown in Equation &14) which was prepared on the basis of ity of the cyclopropane ring opening that denies the radical the Krauss and Smith paper. Thus, the first-formed organo- mechanism [Eq. &12)]. [167] On the basis of a series of cuprate dimer/enone complex with lithium ± carbonyl and control experiments, Bertz et al. reinterpreted the results in terms of CuIII intermediates formed by two-electron O transfer.[168] R Li R R Cu R Cu Cu Li Li O R Li R R Cu R Cu2Me4Li Cu2Me4Li (14) H H D D O R Cu R O O Li Li LiO LiO CuIII (12) R Cu R RR R Me Me "intermediate" rate-determining step Me O Me O H Me [177] H Cu2Me4Li copper ± olefin coordinations forms the product via the D D intermediate&s). A lithium ± carbonyl complex also forms but is a dead-end intermediate. Although the detailed structures 4) Electron spin resonance &ESR) and chemically induced of the intermediates remained obscure for a long time, the dynamic nuclear polarization &CIDNP) spectroscopic essence of this scheme has been supported by subsequent [169] studies for detecting the radical intermediates failed. NMR spectroscopic studies and recent theoretical studies. Conjugate addition of a vinylcuprate reagent to enone The key ªintermediateº is now considered to be an organo- takes place with retention of the vinyl geometry, indicating copper&iii) species formed by two-electron inner-sphere [170, 171] that a vinyl radical intermediate is not involved. electron transfer [Eq. &14)]. [16, 20] Kinetic isotope effects and substituent effects on organo- Corey and Boaz proposed explicitly a Dewar ± Chatt ± cuprate addition to benzophenone indicate that the CÀC Duncanson &DCD) complex for such a CuIII/olefin complex bond formation is rate-determining, which is not consistent &see [Eq. &15)]). [42] X-ray absorption near-edge structure with the involvement of a radical ion pair intermediate.[172] The SET processes do not occur among moderately R electrophilic olefinic acceptors, but are likely to be involved [R Cu]Ð R R R 2 Cu R Cu R with highly electrophilic substrates. A recent example is the HH Cu (15) polyaddition of an organocuprate to fullerenes &see Sec- HH HH H H tion 1.1; [Eq. &4)]). A fluorenone ketyl radical has been DCD complex four-centered TS detected in an organocuprate reaction of fluorenone.[159] Doubly activated olefins[173, 174] and bromonaphthoqui- &XANES) investigation of a complex between a trans- none[175, 176] are also likely to react by SET. cinnamate and Me2CuLi ´ LiI in THF indicated elongation of the CÀCu bond and increase of the coordination number of the copper atom. NMR spectroscopic studies on the organic part of the complexes by Ullenius and co-workers,[178] Krause et al.,[179] and others[180] indicated loosening of the olefinic bond. Very recently, Krause et al. determined, for the first 3.1.3. Kinetics and the Spectroscopic Analysis of À1 time, the kinetic activation energies &Ea 17 ± 18 kcalmol ) Intermediates of some conjugate addition reactions.[181] Conjugate addition to a,b-unsaturated ketones and esters are the most important organocuprate reactions. Important kinetic studies by Krauss and Smith for Me2CuLi and a variety 3.2. Carbocupration of Acetylenes and Olefins of ketones revealed the kinetic expressions shown in Equa- tion &13), which are first order for both the organocuprate Carbocupration of acetylene smoothly takes place in a cis dimer and the enone.[165] fashion to provide a reliable synthetic route to vinyl copper

3760 Angew. Chem. Int. Ed. 2000, 39, 3750 ± 3771 Organocopper Chemistry REVIEWS species [Eq. &15)]. [32] Magnesium and zinc, which are more k3 X Lewis acidic than lithium, are better countercations for this Cu R + CuX reaction, and strong coordination of a lithium dialkylcu- R prate&i) with a crown ether dramatically slows down the reaction.[15] This reaction was generally considered to proceed k2 RR1 + RCu through a four-centered mechanism, and, hence, to be k1 R2Cu 1 X III 1 R R2Cu R mechanistically different from the conjugate addition. + Li // RR + R1Cu In the addition of Me2CuLi reagents to electron-deficient + LiX acetylenes,[182±184] DCD-type complexes have been identified Scheme 8. Two proposed mechanisms for alkylation reactions. by NMR spectroscopy.[185] As shown in Equation &16), ynoates assumes rate-determining displacement of the leaving group X X = OMe R COOMe with copper bearing a formal negative charge, and subsequent [146] Me CuLi¥LiI R formation of a trialkylcopper&iii) intermediate. The latter X 2 O Me Cu R (16) then undergoes reductive elimination to give the cross- O Cu R Me Me Me coupling product. Although the second mechanism may look R = tBu, Me Si X = Me pleasing enough for a copper specialist, it leaves unanswered a 3 DCD complex Me OLi few important questions, namely: the role of the lithium

cation, the relative magnitude of k1 and k2 , and, among others, afford vinylcopper intermediates, and ynones instead afford the reason why exclusive production of a cross-coupled allenolates.[186] The origin of this diversity remains unclear. A product RÀR1 is always observed from the symmetrical 1 III related carbocupration mechanism was also proposed for the R2&R )Cu intermediate. reaction with allenylphosphane oxide.[69] Carbocupration of The proposed participation of a CuIII intermediate is based olefins is known for dienes[187] and cyclopropenes[60] but the on analogy to the chemistry of lithium diorganoaurate&i) I [193±196] III mechanisms also remain unknown. compounds, R2Au Li. Crystallographic data of Cu species[145] further supported the similarity between AuIII and CuIII.[197] The SET mechanism has also been discussed for the 3.3. The S 2 Substitution Reaction on sp3 Carbons N alkylation of secondary alkyl iodides, where the substitution reaction takes place stereorandomly.[29, 191b] The reaction of The SN2 substitution reaction of an alkyl halide with a hard triphenylmethylbromide and Me2CuLi generated an ESR- nucleophile such as an alkyl anion can be most readily active triphenylmethyl radical; this may, however, be regard- achieved with the aid of organocopper chemistry.[188] The S 2 N ed as a special case.[198] Intramolecular cyclization of an reaction with epoxides and aziridines is also synthetically olefinic iodide in the presence of an organocopper reagent has useful.[189] The accelerating effects of BF ´Et O in the latter 3 2 been taken as possible but not conclusive evidence of SET.[199] reaction imply the importance of substrate activation.[190] The alkylation of an alkyl bromide, tosylate, or epoxide with organocuprates takes places with 100% inversion of the configuration at the electrophilic carbon as shown in Equa- 3.4. The SN2'-Allylation Reaction tion &17).[29, 191] The magnitudes of primary and secondary Organocuprates react rapidly with allylic halides &or acetates),[31, 200] propargyl halides &or acetates),[201] and vinyl- Br HPh2CuLi H Ph oxiranes,[30] often with S 2' regioselectivity &Scheme 9). The (17) N Et2O/THF (R) ∆ (S) reaction takes place with anti stereochemistry with respect to 67 %

Ð X [R2Cu] R R + kinetic isotope effects in the reaction of Me2CuLi ´ LiI ´ PBu3 SN2' S 2 with CH3I strongly suggested that the rate-determining step of N [192] the reaction is the SN2-displacement stage. The reaction with alkyl halide, aryl halide, and alkyl tosylate of R2CuLi has Ð X R2Cu III CuIIIR been shown to be first order to the concentration of the X R2Cu 2 [125, 197] &R2CuLi)2 dimer and the alkylating reagent. RCu and [189] Scheme 9. An anti-S 2' allylation reaction with a competing S 2 reaction RCu&PBu3) do not react with epoxides, and alkylation N N pathway. X OAc, halogen, OP&O)Y2. reactions of R2CuLi do not take place in the presence of a crown ether, indicating the importance of a Lewis acidic LiX compound associated with the organocuprate moiety. the leaving group, while syn substitution occurs when an Two mechanistic possibilities have been suggested for the allylic carbamate is employed as the substrate.[202] [5] substitution reactions &Scheme 8). The first assumes simple The reaction of R2CuLi tends to give a mixture of SN2 and

SN2 substitution by the R anion group. The second one SN2' products, which has been suggested to be due to the

Angew. Chem. Int. Ed. 2000, 39, 3750 ± 3771 3761 REVIEWS E. Nakamura and S. Mori involvement of the regioisomeric s-allylic CuIII species shown since the covalent MeÀCu bond with an energy of in brackets in Scheme 9.[200] Studies on substitutent effects in 55 kcalmolÀ1[206] must be cleaved. The neutral RCu species competitive reactions suggested that the rate-determining is, therefore, not a reactive nucleophile. &An ancillary result of stage might involve a two-electron transfer from copper to the this analysis suggests that the conventional ªRCuº reagent [214] À À allylic substrate. The SN2 selectivity of the reaction of must be a certain kind of RCu ´ X cluster, where X may be a

Bu2Cu&X) ´ 2MgBr is higher with X I, OTs than with X halide anion.) &3) Being electron-rich &and having high-lying d [203] Cl, Br, and is also higher in ether than in THF. A orbitals), lithium organocuprates, such as &R2CuLi)2 , bind combination of an organocopper compound and a Lewis tightly to acetylene through two-electron donation from the [49] IV [54] acid, such as RCu ´ BF3,R2CuLi ´ ZnCl2 , R2CuLi ´ Ti , or copper atom &see CP in Scheme 10 and Figure 2). &4) In such [78] R2CuLi ´ AlCl3 , greatly enhances the SN2' selectivity. NMR complex formation, a cluster structure larger than the parent IV spectroscopic studies on R2CuLi ´ ZnCl2 and R2CuLi ´ Ti only species R2CuLi is certainly necessary to achieve cooperation showed rapid transmetalation from Cu to Zn or Ti, giving us of lithium and copper. little information on a putative Cu/Zn or Cu/Ti mixed The reaction pathway may be viewed as a ªtrap-and-biteº reagent.[53, 57] mechanism: The structures are shown in Scheme 10 and Scant information is available for the transition state. The Figure 2, and the approximate energetics in Figure 3. The stereoselectivity of the SN2' reaction of d-substituted allylic cluster opens up and traps the acetylene &INT1), transfers halides suggested that the transition state for the delivery of electrons, and then ªbitesº the substrate to form a CÀC bond an R group from copper has a four-centered character, as &TScc). The important events include: formation of a DCD shown in Equation &18) &S, M, L small, medium, and large complex CP via a low energy transition state TScp,[21] inner- substituents, respectively.[54, 129, 204] This conjecture was supported by theoretical comparison of the transition state geometries of olefin carbolithiation with acetylene carbocup- X X X [16, 205] ration &compare with Scheme 10). oxidative Li Li Li Li Li R Li Li addition X R R R CuIII R I R R I R Cu Cu CuIII + Li H H H H H H HH X M RT TScp CP INT1 M M X [R Cu]Ð S 2 L Cu R L S (18) Li reductive Li transmetalation R H R L elimination X R X R H H I S III CuI Li Cu R Li Cu Li R R X Li H H H H H H PD TScc INT2 4. Theory-Based Molecular Pictures of Scheme 10. The ªtrap and biteº pathway of carbocupration. X RCuR, Organocuprate Reactions halogen, etc.

As summarized in the foregoing sections, numerous experimental studies indicated active participation of large organocopper clusters. They typically bear nucleophilic alkyl residues, copper&i) atoms, and countercations &typically lithium). The Nakamura ± Morokuma theoretical analysis has provided answers to a fair number of questions among the numerous problems in organocuprate chemistry. Three representative reactions, carbocupration, conjugate addition, and SN2 alkylation reaction, have been examined so far in detail to obtain molecular pictures on the cluster participation and the CuI/ CuIII redox chemistry in these reactions.

4.1. The Reaction Pathway of Acetylene Carbocupration

Carbocupration of acetylene was studied systematically for À five model species, MeCu, [Me2Cu] ,Me2CuLi, Me2CuLi ´ [15] LiCl, and &Me2CuLi)2, all of which were invoked once in a while in the discussions of organocuprate mechanisms. A few general conclusions have been made for the reactivities of these reagents with p acceptors. &1) As its d orbital is very low lying &and hence not available for redox chemistry),[21] MeCu Figure 2. ªSnapshotsº of intermediates on the potential surface of can only undergo addition by a four-centered mechanism acetylene carbocupration with Me2CuLi ´ LiCl. Color code: Cu dark [Eq. &5)]. &2) This four-centered path requires high energy, green, Li orange, Cl light green, C large blue, H small blue spheres.

3762 Angew. Chem. Int. Ed. 2000, 39, 3750 ± 3771 Organocopper Chemistry REVIEWS

Note that drawing the ªorganicº arrows and indicating the valence of the metal, as in Scheme 10 &and other schemes in the following paragraphs), are necessarily inaccurate from a purely inorganic or theoretical viewpoint. Nonetheless, we have indicated them to put the theoretical results into the framework of conventional organic chemistry, and to facilitate understanding of the chemistry by organic chemists who are using the reagents in their everyday research life. In Figure 2, snapshots of intermediary species on the potential surface of carbocupration are depicted to illustrate the metamorphosis of the reacting complex. The formation of the transient carbolithiated intermediate INT2 is the most striking because recognition of this intermediate provides the

key to understanding the kinship of carbocupration, SN2'- allylation,[15] and conjugate addition.[16] Figure 3. Approximate energy profile of acetylene carbocupration. The energies are relative to the energies of the reactants. E relative energies, 1 reaction coordinates. &In physical and theoretical chemistry all species on the potential surface are termed intermediates.) 4.2. The Reaction Pathway of Conjugate Addition sphere electron transfer to form a transient intermediate INT1, CÀC bond formation through the rate-determining The reaction pathway of the conjugate addition of Me2CuLi [16] stage &TScc), and intra-cluster transmetalation from lithium and Me2CuLi ´ LiCl has been studied for acrolein and to copper&i) &INT2). The DCD character of CP is shown by the cyclohexenone,[20] and favorably compared for the 13C NMR localized molecular orbitals &LMOs, Figure 4) and was also properties of intermediates, kinetic isotope effects,[207] and the diastereofacial selectivity. The trap-and-bite mechanism also operates in this reaction, as summarized in Scheme 11. As illustrated in Figure 5, the rate-determining step of the reaction &to form TScc) is the CÀC bond formation caused by reductive elimination from CuIII to CuI. Formation of TScc is also the stage where the enantioface selectivity of the reaction is determined.[20] This statement stands in contrast to the conventional assumption that the face selectivity is achieved in the initial p complexation, which is now shown to represent an equilibrium preceding TScc Figure 4. Localized molecular orbitals between Me2CuLi ´ LiCl and acetylene in complex CP. [208] formation. The calculated activation energy &Ea 10.6 kcalmolÀ1), taking into consideration the recently deter- found in the conjugate addition to enals and enones.[16] Since mined solvation of the lithium atoms, shows reasonable III the CÀCu bond is very unstable, the activation energy for agreement with the experimental data &Ea 17 ± the CÀC bond formation via TScc becomes small 18 kcalmolÀ1).[181] &<20 kcalmolÀ1). In solution, the reaction may directly go The central feature of the mechanism is the 3-cuprio&iii) to INT1 or related species through an open cluster, as enolate which is of open dimeric nature. As shown by theory/ depicted in Scheme 5. experiment comparison for 13C NMR spectroscopic results

X R * * I Li Li Li R Li R Li R Li Cu R X R X III X R I R R O Cu O CuIII O Cu Li O III R X O Cu Li Li Li HH HH H H HH HH CPop TScc PD CPcl rate- and stereochemistry-determining step

_ _ R R RR RR Li R I Li R CuIII III Cu O Cu CuI O O X CuI X Li R Li R 3-cuprio(III) enolate cupracyclopropane olefin π−complex

Scheme 11. Possible pathways for conjugate addition of &R2CuLi)2 to an enone. Solvent molecules are omitted for clarity. The lithium atoms are fully solvated. The RÀLi association is indicated with a starred broken line in CPop and TScc but may be extremely small or nonexistent in solution. X RCuR, halogen, etc.

Angew. Chem. Int. Ed. 2000, 39, 3750 ± 3771 3763 REVIEWS E. Nakamura and S. Mori

Synthetic chemists can now work with three-dimensional pictures of conjugate addition which are available on a web site[*] .[20] In the absence of steric hindrance &for example, with 5-methylcyclohexenone), ªaxial attackº through a half-chair conformation is favored &Scheme 12), while ªequatorial attackº through a half-boat conformation is favored under the constraint of bicyclic rings &for example, in the cortisone synthesis in Scheme 3).

Me Me * Li Cu Cu Me O O O Me Li Me H H + (Me CuLi) 2 2 axial attack Figure 5. Approximate energy profile of conjugate addition of &R2CuLi)2 to an enone.

O O OSiMe and kinetic isotope effects,[20, 207] this species serves as a 3 OSiMe3 direct precursor to the product &PD in Scheme 11, upper H H R box). In this critical complex CPop, there is achieved a copper/ + RCu(X)-M+ olefin &soft/soft) interaction and a lithium/carbonyl &hard/ hard) interaction. The open complex may be directly formed Me O through an a open cluster &bottom left of Scheme 11) or OSiMe3 M+ H by complexation of a closed cluster with the enone &to Cu form Cpcl). Experiments have shown that the enone/ X R H lithium complex &top left of Scheme 11) is a dead-end equatorial attack species.[165, 179] Scheme 12. Transition states for diastereoselective conjugate additions. In The intermediate CPop is the ªb-cuprio ketoneº inter- solution, the lithium and metal &M) cations are fully solvated with solvent mediate that has been widely debated in the mechanistic molecules. The MeÀLi association &*) may be extremely weak or nonexistent in solution. X RCuR, halogen, etc. discussions of conjugate addition &compare with Equa- tions &13) and &16)). On the basis of the Nakamura ± Moro- kuma theoretical analysis, one can now consider three limiting structures for CPop that are shown in the lower box in Recognition of the importance of cluster structure led to a Scheme 11. The reason for the exceptional stability of CPop as new understanding of the role of a dummy ligand &Y) in the a trialkylcopper&iii) species can be readily understood with chemistry of mixed cuprates MeCu&Y)Li.[22] As shown in the ª3-cuprio&iii) enolateº structure where the internal Scheme 13 for the case of Y alkynyl, the transfer of the enolate anion acts as a strong stabilizing ligand of the CuIII state.[23]

In spite of the apparent difference between conjugate R addition and carbocupration reactions &Schemes 8 and 9), the X similarity of the key organometallic feature in the two Li Li X Li Me O III R Cu reactions is now evident. In both reactions, inner-sphere + CuI Me Li O electron transfer converts the stable CÀCuI bond to an À III unstable C Cu bond, and the cluster opening generates a R nucleophilic tetracoordinated alkyl group. The difference is + CuI R that the product PD of conjugate addition remains as a X Li LiO CuIII Me lithium enolate complexed with RCuI &Scheme 11), while the Li O initial product of carbocupration &INT2) undergoes further Me Scheme 13. Dummy ligands: selective transfer of the methyl &or alkyl, reaction &Li/Cu transmetalation) and generates a new organo- alkenyl, aryl) group in preference to transfer of the alkynyl group. X cuprate compound &Scheme 10). &Note however that this RCuR, halogen, etc. difference could become more subtle since the product of conjugate addition could behave more like an a-cuprio&i) ketone complexed with a lithium cation than a lithium enolate complexed with copper&i)). In both reactions, no evidence of radical intermediates &that is, from SET) was found by [*]http://www.chem.s.u-tokyo.ac.jp/ common/HTML/3D.structure.html theoretical calculations.[15, 16, 20] for retrieval of the theoretical 3D pictures and coordinates

3764 Angew. Chem. Int. Ed. 2000, 39, 3750 ± 3771 Organocopper Chemistry REVIEWS methyl group is overwhelmingly favored over the transfer of Li X Z Z the alkynyl group. This is because the alkynyl group acts as a X Li R1 Z X tight bridge between CuIII and Li &Figure 6). Namely, the Li Li Li R1 1 Li R R I R Cu R CuI R R CuI R

CP TSsb

oligomers + Solv.

Li Z 1 X Li R RRCuIII Solv. INT1

Figure 6. Three dimensional structure of the open complex between acrolein and Me&C2H)CuLi ´ LiCl, which bears an Me2O group on each 1 lithium atom &B3LYP/631A). Bond lengths are given in angstroms. Color RÐR1 + R 1 code: Cu dark green, Li orange, Cl light green, O red, C dark R III X RRCu blue, H light blue spheres. Li Li RRCuIII Li Li Z R CuI Z Z X X Li Li alkynyl dummy group simultaneously binds to Cu and Li atoms &that is, there is a strong electrostatic interaction PD TScc INT2 1 between the Li atom and alkynyl group), and hence stays on Scheme 14. Reaction of R2CuLi ´ LiX with R Z. The solvent molecules the copper atom. By default, the much less effective bridging coordinated to the lithium atoms are omitted for clarity. Z halogen; X RCuR, halogen, etc. organic ligand is transferred to the enone substrate. This is different to the conventional hypothesis that the Y group forming a stronger CuÀY bond acts as a better dummy ligand &because it resists transfer), providing an additional illustra- tion of the critical roles of cluster structures in organocopper chemistry.

4.3. The Reaction Pathway of the SN2 Alkylation Reaction

The alkylation reaction revealed a different mechanistic aspect of the organocuprate reaction. Theoretical analyses of the reactions of alkyl halides &MeI and MeBr)[17] and epoxides &ethylene oxide and cyclohexene oxide)[18] with lithium organocuprate clusters &&Me2CuLi)2 dimer or Me2CuLi ´ LiCl, Scheme 14 and Figure 7) resolved the long-standing Figure 7. Approximate energy profile of the SN2-substitution reaction between R CuLi ´ LiX and R1Z. questions on the mechanism of the alkylation reaction. The 2 density functional calculations showed that the rate-determining step of the alkylation reaction &through TSsb) is the groups R is secured by the linear geometry of the organo- substitution of the CÀZ bond with an incoming RÀCu s bond. cuprate moiety in the transition state TSsb, which warrants 1 The linear 3dz2 orbital of copper acts here as the nucleophile, cross-coupling between R and R in as shown by the LMO illustrated in Figure 8. The computed TScc. Interestingly, this mechanism and the experimental kinetic isotope effects for the reaction of is a hybrid of the two previous methyl iodide showed good agreement with each other, proposals &shown in Figure 8). supporting this conclusion.[209] A similar reaction pathway was

Note that one can identify again an open-cluster structure found for the SN2 substitution of an in TSsb where the lithium atom electrophilically activates the epoxide with a lithium organocup- leaving group. A trialkylcopper&iii) intermediate may form rate cluster.[18] In contrast to the after the rate-determining, halide-displacement step but only MeBr reaction, the configuration of as an unstable transient species INT1 or INT2 &Scheme 14). the electrophilic carbon center is Figure 8. Localized mo- These are trialkylcopper&iii) complexes of T-shape geometry already inverted in the transition lecular orbitals in the transition state of the S 2 with the fourth ligand &solvent or halide) making a square state, providing the reasoning for N reaction between Me2Cu- planar structure.[23] The trans relationship for the two alkyl the preferred ªtrans diaxial epoxide Li ´ LiCl and MeBr.

Angew. Chem. Int. Ed. 2000, 39, 3750 ± 3771 3765 REVIEWS E. Nakamura and S. Mori openingº that has been widely observed in synthetic studies. 4.5. The Roles of Cluster Structure in Organocuprate

The transition state of the SN2 reaction of cyclohexene oxide is Reactions shown in Equation &19). The previous experimental and theoretical data point out several important characteristics of organocuprate structures Li O and their reaction mechanisms. Cl À À À R1 1) The C Cu C angle in a covalently bonded [R2Cu] R CuLi¥LiCl H O O 2 R1 (19) fragment in stationary states is always nearly 1808.[21] In R1 H Li R ethereal solution, R2CuLi exists as higher aggregates, [127, 136] Cu R whose LiÀR bonds are fractional. One can invariably R identify a neutral fragment, RÀCuÀRÀLi, in crystals of cyclic oligomers and higher polymers &see Scheme 4). Depending on the nature of the reacting electrophiles &s* or p*), either a linear or a bent conformation of the 4.4. Orbital Interactions in Organocuprate Reactions CÀCuÀC moiety becomes important for nucleophilic reactions &Figure 9).[21] The Nakamura ± Morokuma theoretical analysis revealed 2) Due to the fractional RÀLi bonds,[140] clusters and poly- an intriguing difference between the addition reactions and mers can reversibly form open clusters, which trap the unsaturated substrate through multiple-point binding &see the SN2 alkylation reaction for the geometry of the nucleophilic CÀCuÀC moiety. As summarized in Section 2, the Schemes 10 and 11). Lithium cations assist electron flow CÀCuÀC fragment in dicoordinated organocuprate&i) anions from the organocuprate to the electrophile, and to achieve found in stable structures is always linear. As the highest such cooperative action, a cluster of appropriate size may À be necessary. Lewis acidic metals other than lithium &for occupied molecular orbital &HOMO) of a linear [R2Cu] [19, 21] example, ZnII) will also play similar roles. One may molecule is largely the 3dz2 copper orbital, the linear CÀCuÀC group is suitable for interaction with the s* orbital appreciate the elegance in the molecular arrangement in of MeBr, as illustrated in Figure 9a.[21] the reactions of Me2CuLi clusters, as illustrated in Fig- ure 10.

a) b) HOMO HOMO R R R

Cu Cu H H H C

Br Figure 10. Three-dimensional views of the transition state &B3LYP) of:

a) &Me2CuLi)2 with cyclohexenone, b) &Me2CuLi)2 with methyl bromide, σ π *(CÐZ) * (CÐC) and c) Me2CuLi ´ LiCl with ethylene oxide. Color code: Cu dark green, Li orange, Cl light green, O red, Br red with black stripes, C large Figure 9. Orbital interactions between [R Cu]À and substrates in: a) an 2 blue, H small blue spheres. early stage of the reaction with methyl bromide, and b) p-complexation to an acetylene or olefin.

3) A CÀCuI bond is a stable covalent bond, and is difficult to Bending the CÀCuÀC moiety to less than 1508 causes cleave by itself.[213] After charge transfer from organo- mixing of the 3dxz copper orbital with the 2p methyl orbital to cuprate&i) to substrate, however, the cleavage of the make it the HOMO of the organocuprate &Figure 9b), which resulting RÀCuIII bond becomes an easy task. The is now suitable for interaction with the p* orbitals of reductive elimination reaction regenerates RCuI, which enones and acetylenes. Energy gain through back-donation may take part in further catalytic cycles. Thus, in copper- largely compensates the energy loss associated with the catalyzed reactions, excess RÀ anion will react with RCu to bending &approximately 20 kcalmolÀ1 to achieve a 1208 regenerate the necessary organocuprate species. angle). 4) Although acetylene carbocupration and conjugate addi- The above analysis for copper also applies to the same class tion have previously been considered to be two separate of compounds for gold, which, however, forms much more reactions, they have been shown to share essentially the stable CÀAuI bonds[156] and hence is unreactive. On the other same reaction mechanism. The kinship among carbocup- [15] hand, the d orbitals of zinc&ii), a main group neighbor, are too ration, conjugate addition, SN2' allylation, and SN2 low lying to make organozinc compounds as nucleophilic as alkylation has now been established, through the Naka- organocopper compounds.[19] mura ± Morokuma theoretical studies.

3766 Angew. Chem. Int. Ed. 2000, 39, 3750 ± 3771 Organocopper Chemistry REVIEWS

5) Demonstration of the critical role of the open conforma- Y. Yamamoto, Angew. Chem. 1986, 98, 945 ± 957; Angew. Chem. Int. tion of the polymetallic clusters highlights the theoretical Ed. Engl. 1986, 25, 947 ± 959. analysis of the organocuprate chemistry. Polymetallic [4]E. Nakamura, Synlett 1991, 539 ± 547. [5]B. H. Lipshutz, S. Sengupta, Org. React. 1992, 41, 135 ± 631. clusters in various synthetic reactions are currently attract- [6]a) Organocopper Reagents &Ed.: R. J. K. Taylor), Oxford University ing the attention of synthetic and mechanistic chemists Press, Oxford, 1994; b) B. H. Lipshutz in Comprehensive Organo- alike.[210±216] metallic Chemistry II, Vol. 12 &Eds.: E. W. Abel, F. G. A. Stone, G. Wilkinson), Pergamon, Oxford, 1995, pp. 59 ± 130; c) N. Krause, A. Gerold, Angew. Chem. 1997, 109, 194 ± 213; Angew. Chem. Int. Ed. Engl. 1997, 36, 186 ± 204; d) Y. Ibuka, Y. Yamamoto, Synlett 1992, 5. Summary and Outlook 769 ± 777. [7]a) G. H. Posner, Org. React. 1972, 19, 1 ± 113; b) J. A. Kozlowski in Comprehensive Organic Synthesis, Vol. 4 &Eds.: B. M. Trost, I. Wherefore art thou copper? The uniqueness of the organo- Fleming), Pergamon, Oxford, 1991, p. 169 ± 198; c) P. Perlmutter, copper chemistry stems primarily from the fact that copper Conjugate Addition Reactions in Organic Synthesis, Pergamon, lies on the border line between main group and transition Oxford, 1992. metal elements. Comparison may be made for the neighbor- [8]J. F. Normant, A. Alexakis, Synthesis 1981, 841 ± 870. ing elements, Ni0,CuI,AgI,AuI, and ZnII, all of which are in [9]a) G. H. Posner, Org. React. 1975, 22, 253 ± 400; b) J. M. Klunder, G. H. Posner in Comprehensive Organic Synthesis, Vol. 3 &Eds.: B. M. 10 the d configuration. The energy levels of the 3d orbitals in Trost, I. Fleming), Pergamon, Oxford, 1991, pp. 207 ± 239. copper&i) compounds are much higher than in zinc&ii), and [10]J. A. Marshall, Chem. Rev. 1989, 89, 1503 ± 1511. become even higher upon mixing with the 2p orbital of the [11]M. S. Kharasch, P. O. Tawney, J. Am. Chem. Soc. 1941, 63, 2308 ± alkyl ligand through [R Cu]À formation.[15, 21] A redox system 2315. 2 [12]J. P. Collman, L. S. Hegedus, J. R. Norton, R. G. Finke, Principles and such as the CuI/CuIII system is unavailable for zinc&ii). Applications of Organotransition Metal Chemistry, 2nd ed., Univer- Organonickel and organosilver species are not as stable and sity Science Books, Mill Valley, CA, 1987, Chap. 14. are hence much less synthetically viable than organocopper&i) [13]a) A. E. Jukes, Adv. Organomet. Chem. 1974, 12, 215 ± 322; G. reagents. Organogold&i) species are too stable to be syntheti- van Koten, J. G. Noltes in Comprehensive Organometallic Chemistry, cally useful. The CÀCuÀC angle is intimately connected to the Vol. 2 &Eds.: G. Wilkinson, F. G. A. Stone), Pergamon, Oxford, 1982, pp. 709 ± 763; b) G. van Koten, S. L. James, J. T. B. H. Jastrzebski in reactivities of the diorganocuprate&i), and the Lewis acid Comprehensive Organometallic Chemistry II, Vol. 3 &Eds.: E. W. &often Li) in organocuprate clusters provides a push ± pull Abel, F. G. A. Stone, G. Wilkinson), Pergamon, Oxford, 1995, electronic assistance for charge transfer from CuI to the pp. 57 ± 133; c) P. P. Power, Prog. Inorg. Chem. 1991, 39, 75 ± 11. electrophile. The diversity of the coordination structures [14]R. A. J. Smith, A. S. Vellekoop in Advances in Detailed Reaction Mechanisms, Vol. 3 &Ed.: J. M. Coxon), JAI, Greenville, CT, 1994, revealed by calculations indicates that organocopper chem- pp. 79 ± 130. istry represents an ultimate ªsupramolecular chemistryº that [15]E. Nakamura, S. Mori, M. Nakamura, K. Morokuma, J. Am. Chem. chemists have exploited for a long time without knowing it. Soc. 1997, 119, 4887 ± 4899. Numerous other aspects of organocopper chemistry await [16]E. Nakamura, S. Mori, K. Morokuma, J. Am. Chem. Soc. 1997, 119, further mechanistic studies. The importance of R CuIII species 4900 ± 4910. 3 [17]E. Nakamura, S. Mori, K. Morokuma, J. Am. Chem. Soc. 1998, 120, [23] is now fully recognized, and needs more careful attention in 8273 ± 8274. the future studies of mechanistic and synthetic organocopper [18]S. Mori, E. Nakamura, K. Morokuma, J. Am. Chem. Soc. 2000, 122, chemistry. 7294 ± 7307. [19]S. Mori, E. Nakamura, J. Mol. Struct. 4Theochem) 1999, 461± 462 , 167 ± 175. We thank Prof. Keiji Morokuma for fruitful collaboration on [20]S. Mori, E. Nakamura, Chem. Eur. J. 1999, 5, 1534 ± 1543. the theoretical analysis of copper chemistry, and Monbusho [21]S. Mori, E. Nakamura, Tetrahedron Lett. 1999, 40, 5319 ± 5322; S. 4The Ministry of Education, Science, Culture, and Sport), Mori, A. Hirai, M. Nakamura, E. Nakamura, Tetrahedron 2000, 56, Japan for financial support in the form of a Grant-in-Aid for 2805 ± 2807. Scientific Research on a Priority Area 4No. 283, Innovative [22]E. Nakamura, M. Yamanaka, J. Am. Chem. Soc. 1999, 121, 8941 ± 8942. [23]E. Nakamura, M. Yamanaka, S. Mori, J. Am. Chem. Soc. 2000, 122, Synthetic Reactions). A part of the work was carried out under 1826 ± 1827. the JSPS/NSF Cooperative Research Program. E.N. thanks the [24]J. Munch-Petersen, J. Org. Chem. 1957, 22, 170 ± 176. Colorado State University for the Syntex Distinguished Lec- [25]H. Gilman, R. G. Jones, L. A. Woods, J. Org. Chem. 1952, 17, 1630 ± tureship, which stimulated the idea of writing this review. S.M. 1634. [26]G. Costa, A. Camus, L. Gatti, N. Marsich, J. Organomet. Chem. 1966, is grateful for a JSPS postdoctoral fellowship. 5, 568 ± 572. [27]H. O. House, W. L. Respess, G. M. Whitesides, J. Org. Chem. 1966, Received: August 18, 1999 31, 3128 ± 3141. Revised: February 14, 2000 [A358] [28]E. J. Corey, G. H. Posner, J. Am. Chem. Soc. 1967, 89, 3911 ± 3912; E. J. Corey, G. H. Posner, J. Am. Chem. Soc. 1968, 90, 5615 ± 5616. [29]G. M. Whitesides, W. F. Fischer, Jr., J. San Filippo, Jr., R. W. Bashe, [1]a) J. G. Bednorz, K. A. Müller, Z. Phys. B 1986, 64, 189 ± 193; b) L. H. O. House, J. Am. Chem. Soc. 1969, 91, 4871 ± 4882. Gao, Y. Y. Xue, F. Chen, Q. Xiong, R. L. Meng, D. Ramirez, C. W. [30]R. Anderson, C. A. Henrick, J. B. J. Siddall, J. Am. Chem. Soc. 1970, Chu, J. H. Eggert, H. K. Mao, Phys. Rev. B 1994, 50, 4260. 92, 735 ± 737; R. J. Anderson, J. Am. Chem. Soc. 1970, 92, 4978 ± [2]a) S. J. Lippard, Science 1995, 268, 996 ± 997; b) N. Kitajima, Y. Moro- 4980; R. W. Herr, C. R. Johnson, J. Am. Chem. Soc. 1970, 92, 4980 ± oka, Chem. Rev. 1994, 94, 737 ± 757; c) M. J. Henson, P. Mukherjee, 4981; J. Staroscik, B. Rickborn, J. Am. Chem. Soc. 1971, 93, 3046 ± D. E. Root, T. D. P. Stack, E. I. Solomon, J. Am. Chem. Soc. 1999, 3047; D. M. Wieland, C. R. Johnson, J. Am. Chem. Soc. 1971, 93, 121, 10332 ± 10 345, and references therein. 3047 ± 3049. [3]J. F. Normant, Synthesis 1972, 63 ± 80; G. H. Posner, An Introduction [31]Compare with: H. L. Goering, S. Kantner, J. Org. Chem. 1981, 46, to Synthesis Using Organocopper Reagents, Wiley, New York, 1980; 2144 ± 2148.

Angew. Chem. Int. Ed. 2000, 39, 3750 ± 3771 3767 REVIEWS E. Nakamura and S. Mori

[32]A. Alexakis, J. Normant, J. Villie Âras, Tetrahedron Lett. 1976, 3461 ± [64]a) B. L. Feringa, M. Pineschi, L. A. Arnold, R. Imbos, A. H. M. 3462. de Vries, Angew. Chem. 1997, 109, 2733 ± 2736; Angew. Chem. Int. [33]B. H. Lipshutz, J. A. Kozlowski, C. M. Breneman, J. Am. Chem. Soc. Ed. Engl. 1997, 36, 2620 ± 2623; b) A. K. H. Knöbel, I. H. Escher, A. 1985, 107, 3197 ± 3204. Pfaltz, Synlett 1997, 1429 ± 1431; c) R. Naasz, L. A. Arnold, M. [34]a) T. Mukaiyama, K. Narasaka, K. Banno, Chem. Lett. 1973, 1011 ± Pineschi, E. Keller, B. L. Feringa, J. Am. Chem. Soc. 1999, 121, 1104 ± 1014; b) T. Mukaiyama, K. Banno, K. Narasaka, J. Am. Chem. Soc. 1105. 1974, 96, 7503 ± 7509. [65]M. Sawamura, H. Iikura, E. Nakamura, J. Am. Chem. Soc. 1996, 118, [35]a) K. Maruyama, Y. Yamamoto, J. Am. Chem. Soc. 1977, 99, 8068 ± 12850 ± 12 851. 8070; b) Y. Yamamoto, K. Maruyama, J. Am. Chem. Soc. 1978, 100, [66]M. Sawamura, H. Iikura, A. Hirai, E. Nakamura, J. Am. Chem. Soc. 3240 ± 3241. 1998, 120, 8285 ± 8286. [36]A. B. Smith III, P. J. Jerris, J. Am. Chem. Soc. 1981, 103, 194 ± 195. [67]H. Iikura, S. Mori, M. Sawamura, E. Nakamura, J. Org. Chem. 1997, [37]B. H. Lipshutz, E. L. Ellsworth, S. H. Dimock, J. Am. Chem. Soc. 62, 7912 ± 7913. 1990, 112, 5869 ± 5871. [68]H. O. House, Acc. Chem. Res. 1976, 9, 59 ± 67. [38]Y. Horiguchi, E. Nakamura, I. Kuwajima, J. Am. Chem. Soc. 1989, [69]J. Berlan, J.-P. Battioni, K. Koosha, Tetrahedron Lett. 1976, 3351 ± 111, 6257 ± 6265. 3354; K. Koosha, J. Berlan, M.-L. Capmau, W. Chodkiewicz, Bull. [39]E. Nakamura, I. Kuwajima, J. Am. Chem. Soc. 1984, 106, 3368± Chim. Soc. Fr. II 1975, 1284 ± 1291; K. Koosha, J. Berlan, M.-L. 337. Capmau, W. Chodkiewicz, Bull. Chim. Soc. Fr. 1975, 1291 ± 1294; J. [40]J. Enda, I. Kuwajima, J. Am. Chem. Soc. 1985, 107, 5495 ± 550. Berlan, J.-P. Battioni, K. Koosha, Bull. Chim. Soc. Fr. 1978, II183 ± [41]E. Nakamura, S. Matsuzawa, Y. Horiguchi, I. Kuwajima, Tetrahedron 190; J. Berlan, J.-P. Battioni, K. Koosha, J. Organomet. Chem. 1978, Lett. 1986, 27, 4029 ± 4032. 152, 359 ± 365; J. Berlan, K. Koosha, J. Organomet. Chem. 1978, 153, [42]E. J. Corey, N. W. Boaz, Tetrahedron Lett., 1985, 26, 6015 ± 6018; E. J. 99 ± 106; J. Berlan, K. Koosha, J. Organomet. Chem. 1978, 153, 107 ± Corey, N. W. Boaz, Tetrahedron Lett., 1985, 26, 6019 ± 6022. 113. [43]A. Alexakis, J. Berlan, Y. Besace, Tetrahedron Lett. 1986, 27, 1047 ± [70]E. J. Corey, D. J. Beames, J. Am. Chem. Soc. 1972, 94, 7210 ± 7211. 1050. [71]a) E. J. Corey, D. M. Floyd, B. H. Lipshutz, J. Org. Chem. 1978, 43, [44]Y. Horiguchi, S. Matsuzawa, E. Nakamura, I. Kuwajima, Tetrahedron 3418 ± 3420; b) H. O. House, M. J. Umen, J. Org. Chem. 1973, 38, Lett. 1986, 27, 4025 ± 4028; S. Matsuzawa, Y. Horiguchi, E. Naka- 3893 ± 3901; c) W. H. Mandeville, G. M. Whitesides, J. Org. Chem. mura, I. Kuwajima, Tetrahedron 1989, 45, 349 ± 362. 1974, 39, 400 ± 405. [45]R. J. Linderman, A. Godfrey, Tetrahedron Lett. 1986, 27, 4553. [72]a) J. P. Gorlier, L. Hamon, J. Levisalles, J. Wagnon, J. Chem. Soc. [46]E. Nakamura in Organocopper Reagents &Ed.: R. J. K. Taylor), Chem. Commun. 1973, 88; b) B. H. Lipshutz, R. S. Wilhelm, J. A. Oxford University Press, Oxford, 1994, Chap. 6, pp. 129 ± 142. Kozlowski, Tetrahedron 1984, 40, 5005 ± 5038. [47]E. Nakamura, S. Aoki, H. Oshino, I. Kuwajima, J. Am. Chem. Soc. [73]S. H. Bertz, G. Dabbagh, G. M. Villacorta, J. Am. Chem. Soc. 1982, 1985, 109, 8056 ± 8066. 104, 5824 ± 5826; compare with: G. H. Posner, C. E. Whiten, Tetra- [48]Reviews: a) E. Nakamura, J. Synth. Org. Chem. Jpn. 1989, 47, 931 ± hedron Lett. 1973, 1815 ± 1818; G. H. Posner, C. E. Whiten, J. J. 938; b) ªMetal Homoenolatesº: I. Kuwajima, E. Nakamura in Sterling, J. Am. Chem. Soc. 1973, 95, 7788 ± 7800. Comprehensive Organic Synthesis, Vol. 2 &Eds.: B. M. Trost, I. [74]Magnesium mixed organocuprates: J. Drouin, F. Leyendecker, J.-M. Fleming), Pergamon, Oxford, 1991, Chap. 1.14; c) M. Nakamura, E. Conia, New. J. Chem. 1978, 2, 267 ± 270; F. Leyendecker, J. Drouin, J.- Nakamura, J. Synth. Org. Chem. Jpn. 1998, 56, 632 ± 644. M. Conia, New. J. Chem. 1978, 2, 271 ± 274. [49]E. Nakamura, I. Kuwajima, J. Am. Chem. Soc. 1977, 99, 7360 ± 7361; [75]C. Ouannes, G. Dressaire, Y. Langlois, Tetrahedron Lett. 1977, 815 ± E. Nakamura, H. Oshino, I. Kuwajima, J. Am. Chem. Soc. 1986, 108, 818. 3745 ± 3755. [76]S. H. Bertz, C. P. Gibson, G. Dabbagh, Organometallics 1988, 7, 227 ± [50]a) Y. Tamaru, H. Ochiai, T. Nakamura, Z. Yoshida, Angew. Chem. 232. 1987, 99, 1193 ± 1195; Angew. Chem. Int. Ed. Engl. 1987, 26, 1157 ± [77]A. Alexakis, D. Jachiet, J. F. Normant, Tetrahedron 1986, 42, 5607 ± 1159; b) Y. Tamaru, H. Ochiai, F. Sanda, Z.-i. Yoshida, Tetrahedron 5619; A. Ghribi, A. Alexakis, J. F. Normant, Tetrahedron Lett. 1984, Lett. 1985, 26, 5529 ± 5532; c) Y. Tamaru, H. Ochiai, T. Nakamura, K. 25, 3075 ± 3078; A. Ghribi, A. Alexakis, J. F. Normant, Tetrahedron Tsubaki, Z. Yoshida, Tetrahedron Lett. 1985, 26, 5559 ± 5562. Lett. 1984, 25, 3079 ± 3082. [51]a) P. Knochel, M. C. P. Yeh, S. C. Berk, J. Talbert, J. Org. Chem. 1988, [78]S. Flemming, J. Kabbara, K. Nickisch, J. Westermann, J. Mohr, 53, 2390; b) P. Knochel, Synlett 1995, 393 ± 403. Synlett 1995, 183 ± 185. [52]P. Knochel, R. D. Singer, Chem. Rev. 1993, 93, 2117 ± 2188. [79]S. Matsuzawa, M. Isaka, E. Nakamura, I. Kuwajima, Tetrahedron [53]E. Nakamura in Organometallic Chemistry, A Manual, 2nd ed. &Ed.: Lett. 1989, 30, 1975 ± 1978; M. Arai, T. Nemoto, Y. Ohashi, E. M. Schlosser), Wiley, New York, in preparation. Nakamura, Synlett 1992, 309 ± 310. [54]M. Arai, E. Nakamura, B. H. Lipshutz, J. Org. Chem. 1991, 56, 5489 ± [80]For some mechanistic proposals, see: a) B. H. Lipshutz, S. H. 5491. Dimock, B. James, J. Am. Chem. Soc. 1993, 115, 9283 ± 9284; [55]B. H. Lipshutz, E. L. Ellsworth, J. Am. Chem. Soc. 1990, 112, 7440 ± b) S. H. Bertz, G. Miao, B. E. Rossiter, J. P. Snyder, J. Am. Chem. 7441. Soc. 1995, 117, 11023 ± 11 024; c) S. H. Bertz, A. Chopra, M. Eriksson, [56]B. H. Lipshutz, E. L. Ellsworth, S. H. Dimock, J. Am. Chem. Soc. C. A. Ogle, P. Seagle, Chem. Eur. J. 1999, 5, 2680 ± 2691. 1990, 112, 5869 ± 5871. [81]Y. Horiguchi, M. Komatsu, I. Kuwajima, Tetrahedron Lett. 1989, 30, [57]T. Ibuka, H. Minakata, Synth. Commun. 1980, 10, 119 ± 125. 7087 ± 7090. [58]M. Arai, B. H. Lipshutz, E. Nakamura, Tetrahedron 1992, 48, 5709 ± [82]B. H. Lipshutz, D. H. Aue, B. James, Tetrahedron Lett. 1996, 37, 5718. 8471 ± 8474. [59]A. Alexakis, J. F. Normant, Isr. J. Chem. 1984, 24, 113 ± 117. [83]M. Eriksson, A. Johansson, M. Nilsson, T. Olsson, J. Am. Chem. Soc. [60]A. T. Stoll, E.-i. Negishi, Tetrahedron Lett. 1985, 26, 5671 ± 5674. 1996, 118, 10904 ± 10905. [61]J. Berlan, J. Besace, E. Stephan, P. Cresson, Tetrahedron Lett. 1985, [84]M. Bergdahl, E.-L. Lindstedt, M. Nilsson, T. Olsson, Tetrahedron 26, 5765 ± 5768. 1988, 44, 2055 ± 2062. [62]E. Nakamura, M. Isaka, S. Matsuzawa, J. Am. Chem. Soc. 1988, 110, [85]a) E. C. Ashby, J. Lin, J. J. J. Watkins, J. Org. Chem. 1977, 42, 1099 ± 1297 ± 1298. 1102; b) E. C. Ashby, J. J. Lin, J. Org. Chem. 1977, 42, 2805 ± 2808. [63]a) B. E. Rossiter, N. M. Swingle, Chem. Rev. 1992, 92, 771 ± 806; b) T. [86]a) H. Westmijze, H. Kleijn, J. Meijer, P. Vermeer, Recl. Trav. Chim. Kanai, Y. Nakagawa, K. Tomioka, J. Syn. Org. Chem. Jpn. 1996, 35, Pays-Bas 1981, 100, 98 ± 102; b) D. L. J. Clive, V. Farina, P. L. 474 ± 480; c) G. van Koten, Pure Appl. Chem. 1994, 66, 1455 ± 1462; Beaulieu, J. Org. Chem. 1982, 47, 2572 ± 2582. d) A. Alexakis, J. C. Frutos, P. Mangeney, Tetrahedron: Asymmetry [87]T. L. Macdonald, W. C. Still, J. Am. Chem. Soc. 1975, 97, 5280 ± 5281; 1993, 4, 2427 ± 2430; e) A. Alexakis in Transition Metal Catalysed W. C. Still, T. L. Macdonald, Tetrahedron Lett. 1976, 31, 2659 ± 2662. Reactions &Eds.: S.-I. Murahashi, S. G. Davies), Blackwell Science, [88]E. C. Ashby, J. J. Watkins, J. Am. Chem. Soc. 1977, 99, 5312 ± Oxford, UK, 1999, pp. 303. 5317.

3768 Angew. Chem. Int. Ed. 2000, 39, 3750 ± 3771 Organocopper Chemistry REVIEWS

[89]S. H. Bertz, G. Dabbagh, J. Am. Chem. Soc. 1988, 110, 3668 ± 3670. [118]M. M. Olmstead, P. P. Power, J. Am. Chem. Soc. 1990, 112, 8008 ± [90]M. M. Olmstead, P. P. Power, J. Am. Chem. Soc. 1989, 111, 4135 ± 8014. 4136. [119]C.-S. Hwang, P. P. Power, Organometallics 1999, 18, 697 ± 700. [91]F. Olbrich, J. Kopf, E. Weiss, Angew. Chem. 1993, 105, 1136; Angew. [120]H. Hope, M. M. Olmstead, P. P. Power, J. Sandell, X. Xu, J. Am. Chem. Int. Ed. Engl. 1993, 32, 1077 ± 1079. Chem. Soc. 1985, 107, 4337 ± 4338. [92]B. H. Lipshutz, R. S. Wilhelm, J. A. Kozlowski, Synthesis 1987, 325 ± [121]a) J. P. Snyder, G. E. Tipsword, D. J. Spangler, J. Am. Chem. Soc. 341; B. H. Lipshutz, R. S. Wilhelm, D. M. Floyd, J. Am. Chem. Soc. 1992, 114, 1507 ± 1510; b) See also Ref. [99a]. 1981, 103, 7672 ± 7674; B. H. Lipshutz, J. Kozlowski, R. S. Wilhelm, J. [122]S. F. Martin, J. R. Fishpaugh, J. M. Power, D. M. Giolando, R. A. Am. Chem. Soc. 1982, 104, 2305 ± 2307; B. H. Lipshutz, Tetrahedron Jones, C. M. Nunn, A. H. Cowley, J. Am. Chem. Soc. 1988, 110, Lett. 1983, 24, 127 ± 130; B. H. Lipshutz, K. Siegmann, E. Garcia, F. 7226 ± 7228. Kayser, J. Am. Chem. Soc. 1993, 115, 9276 ± 9282. [123]C.-S. Hwang, P. P. Power, J. Am. Chem. Soc. 1998, 120, 6409 ± 6410. [93]S. H. Bertz, J. Am. Chem. Soc. 1990, 112, 4031 ± 4032; S. H. Bertz, J. [124]G. van Koten, J. G. Noltes, J. Chem. Soc. Chem. Commun. 1972, Am. Chem. Soc. 1991, 113, 5470 ± 5471; S. H. Bertz, Chem. Eng. News 940 ± 941. 1997, 75&15), 4. [125]R. G. Pearson, C. D. Gregory, J. Am. Chem. Soc. 1976, 98, 4098 ± [94]S. H. Bertz, K. Nilsson, OÃ . Davidson, J. P. Snyder, Angew. Chem. 4104. 1998, 110, 327 ± 331; Angew. Chem. Int. Ed. 1998, 37, 314 ± 317. [126]G. van Koten, J. T. B. H. Jastrzebski, F. Muller, C. H. Stam, J. Am. [95]B. H. Lipshutz, S. Sharma, E. L. Ellsworth, J. Am. Chem. Soc. 1990, Chem. Soc. 1985, 107, 697 ± 698. 112, 4032 ± 4034; B. H. Lipshutz, B. James, J. Org. Chem. 1994, 59, [127]N. P. Lorenzen, E. Weiss, Angew. Chem. 1990, 102, 322 ± 324; Angew. 7585 ± 7587. Chem. Int. Ed. Engl. 1990, 29, 300 ± 302. [96]a) T. Stemmler, J. E. Penner-Hahn, P. Knochel, J. Am. Chem. Soc. [128]M. M. Olmstead, P. P. Power, Organometallics 1990, 9, 1720 ± 1993, 115, 348 ± 350; b) T. M. Barnhart, H. Huang, J. E. Penner- 1722. Hahn, J. Org. Chem. 1995, 60, 4310 ± 4311. [129]3D structures searched by Cambridge Structural Database System in [97]T. L. Stemmler, T. M. Barnhart, J. E. Penner-Hahn, C. E. Tucker, P. April, 1999. Knochel, M. Böhme, G. Frenking, J. Am. Chem. Soc. 1995, 117, [130]H. Huang, C. H. Liang, J. E. Penner-Hahn, Angew. Chem. 1998, 110, 12489 ± 12 497. 1628 ± 1630; Angew. Chem. Int. Ed. 1998, 37, 1564 ± 1566. [98]B. H. Lipshutz, K. L. Stevens, B. James, J. G. Pavlovich, J. P.Snyder, J. [131]S. I. Khan, P. G. Edwards, H. S. H. Yuan, R. Bau, J. Am. Chem. Soc. Am. Chem. Soc. 1996, 118, 6796 ± 6797; B. H. Lipshutz, J. Keith, D. J. 1985, 107, 1682 ± 1684. Buzard, Organometallics 1999, 18, 1571 ± 1574. [132]E. Nakamura, J. Synth. Org. Chem. Jpn. 1993, 51, 985 ± 994. [99]a) J. P. Snyder, D. P. Spangler, J. R. Behling, B. E. Rossiter, J. Org. [133]B. H. Lipshutz, J. Keith, D. J. Buzard, Organometallics 1999, 18, Chem. 1994, 59, 2665 ± 2667; b) J. P. Snyder, S. H. Bertz, J. Org. 1571 ± 1574. Chem. 1995, 60, 4312 ± 4313; c) H. Huang, K. Alvarez, Q. Cui, T. M. [134]T. M. Barnhart, H. Huang, J. E. Penner-Hahn, J. Org. Chem. 1995, 60, Barnhart, J. P. Snyder, J. E. Penner-Hahn, J. Am. Chem. Soc. 1996, 4310 ± 4311. 118, 8808 ± 8816; erratum: H. Huang, K. Alvarez, Q. Cui, T. M. [135]A. Gerold, J. T. B. H. Jastrzebski, C. M. P. Kronenburg, N. Krause, G. Barnhart, J. P. Snyder, J. E. Penner-Hahn, J. Am. Chem. Soc. 1996, van Koten, Angew. Chem. 1997, 109, 778 ± 780; Angew. Chem. Int. 118, 12 252. Ed. Engl. 1997, 36, 755 ± 757. [100]G. Boche, F. Bosold, M. Marsch, K. Harms, Angew. Chem. 1998, 110, [136]T. Stemmler, J. E. Penner-Hahn, P. Knochel, J. Am. Chem. Soc. 1993, 1779 ± 1781; Angew. Chem. Int. Ed. 1998, 37, 1684 ± 1686. 115, 348 ± 350. [101]C. M. P. Kronenburg, J. T. B. H. Jastrzebski, A. L. Spek, G. van Ko- [137]S. H. Bertz, A. S. Vellekoop, R. A. J. Smith, J. P. Snyder, Organo- ten, J. Am. Chem. Soc. 1998, 120, 9688 ± 9689. metallics 1995, 14, 1213 ± 1220. [102]S. H. Bertz, G. Dabbagh, J. Chem. Soc. Chem. Commun. 1982, 1030 ± [138]M. Böhme, G. Frenking, M. T. Reetz, Organometallics 1994, 13, 1032. 4237 ± 4245. [103]S. H. Bertz, G. Miao, M. Eriksson, Chem. Commun. 1996, 815 ± 816. [139]G. Fraenkel, M. Henrichs, J. M. Hewitt, B. M. Su, M. J. Geckele, J. [104]T. A. Mobley, F. Müller, S. Berger, J. Am. Chem. Soc. 1998, 120, Am. Chem. Soc. 1980, 102, 3345 ± 3350. 1333 ± 1334. [140]J. Heinzer, J. F. M. Oth, D. Seebach, Helv. Chim. Acta 1985, 68, [105]N. Krause, Angew. Chem. 1999, 111, 83 ± 85; Angew. Chem. Int. Ed. 1848 ± 1862. 1999, 38, 79 ± 81. [141]G. Hallnemo, C. Ullenius, Tetrahedron 1983, 39, 1621 ± 1625. [106]J. A. Cabezas, A. C. Oehlschlager, J. Am. Chem. Soc. 1997, 119, [142]C. L. Kingsbury, R. A. J. Smith, J. Org. Chem. 1997, 62, 4629 ± 4634; 3878 ± 3886. C. L. Kingsbury, R. A. J. Smith, J. Org. Chem. 1997, 62, 7637 ± 7643. [107]W. C. Still, J. A. Schneider, Tetrahedron Lett. 1980, 21, 1035 ± 1038. [143]C. L. Kingsbury, K. S. Sharp, R. A. J. Smith, Tetrahedron 1999, 55, [108]G. M. Whitesides, J. SanFilippo, Jr., C. P. Casey, E. J. Panek, J. Am. 14693 ± 14 700. Chem. Soc. 1967, 89, 5302 ± 5303. [144]C. Kappenstein, R. P. Hugel, Inorg. Chem. 1978, 17, 1945 ± 1949; P. [109]G. H. Posner, J.-S. Ting, C. M. Lentz, Tetrahedron 1976, 32, 2281 ± Chaudhuri, K. Oder, K. Wieghardt, J. Weiss, J. Reedijk, V. Hinrichs, J. 2287. Wood, A. Ozarowski, H. Stratemaier, D. Reinen, Inorg. Chem. 1986, [110]P. Kocienski, S. Wadman, K. Cooper, J. Am. Chem. Soc. 1989, 111, 25, 2951 ± 2958; J. Cernak, J. Chomic, C. Kappenstein, M. Dunaj- 2363 ± 2365. Jurco, Z. Kristallogr. 1994, 209, 43. [111]G. V. Buxton, J. C. Green, J. Chem. Soc. Faraday Trans. 1978, 74, [145]a) M. A. Willert-Porada, D. J. Burton, N. C. Baenziger, J. Chem. Soc. 697 ± 714. Chem. Commun. 1989, 1633 ± 1634; b) D. Naumann, T. Roy, K.-F. [112]J. K. Kochi, Acc. Chem. Res. 1974, 7, 351 ± 360; compare with: I. Ryu, Tebbe, W. Crump, Angew. Chem. 1993, 105, 1555 ± 1556; Angew. M. Ando, A. Ogawa, S. Murai, N. Sonoda, J. Am. Chem. Soc. 1983, Chem. Int. Ed. Engl. 1993, 32, 1482 ± 1483; c) R. Eujen, B. Hoge, D. J. 105, 7192 ± 7194. Brauer, J. Organomet. Chem. 1996, 519, 7 ± 20. [113]Metastable dimethylgold& ii) clusters were recently trapped. See: D. [146]a) A. E. Dorigo, J. Wanner, P. von R. Schleyer, Angew. Chem. 1995, Zhu, S. V. Lindeman, J. K. Kochi, Organometallics 1999, 18, 2241 ± 107, 492 ± 494; Angew. Chem. Int. Ed. Engl. 1995, 34, 476 ± 478; b) J. P. 2248. Snyder, J. Am. Chem. Soc. 1995, 117, 11025 ± 11026. [114]P. S. Coan, K. Folting, J. C. Huffman, G. Caulton, Organometallics [147]J. P. Snyder, Angew. Chem. 1995, 107, 112 ± 113; Angew. Chem. Int. 1989, 8, 2724 ± 2728; R. Lingnau, J. Strähle, Angew. Chem. 1988, 100, Ed. Engl. 1995, 34, 80 ± 81. 409; Angew. Chem. Int. Ed. Engl. 1988, 27, 436. [148]M. Kaupp, H. G. von Schnering, Angew. Chem. 1995, 107, 1076; [115]S. Sakaki, Y. Musashi, Inorg. Chem. 1995, 34, 1914 ± 1923. Angew. Chem. Int. Ed. Engl. 1995, 34, 986. [116]M. D. Janssen, M. A. Corsten, A. L. Spek, D. M. Grove, G. van Ko- [149]J. P. Snyder, Angew. Chem. 1995, 107, 1076 ± 1077; Angew. Chem. Int. ten, Organometallics 1996, 15, 2810 ± 2820. Ed. Engl. 1995, 34, 986 ± 987; compare with: J. K. Bera, A. G. [117]M. F. Lappert, R. Pearce, J. Chem. Soc. Chem. Commun. 1973,24± Samuelson, J. Chandrasekhar, Organometallics 1998, 17, 4136 ± 4145. 25; J. A. J. Jarvis, R. Pearce, M. F. Lappert, J. Chem. Soc. Dalton [150]R. A. Poirier, G. A. Ozin, D. F. Mcintosh, G. Csizmadia, R. Daudel, Trans. 1977, 999 ± 1003. Chem. Phys. Lett. 1983, 191, 221 ± 229.

Angew. Chem. Int. Ed. 2000, 39, 3750 ± 3771 3769 REVIEWS E. Nakamura and S. Mori

[151]T. Ziegler, V. Tschinke, A. Becke, J. Am. Chem. Soc. 1987, 109, 1351 ± [184]E. J. Corey, J. A. Katzenellenbogen, J. Am. Chem. Soc. 1969, 91, 1358. 1851 ± 1852; J. B. Siddall, M. Biskup, J. H. Fried, J. Am. Chem. Soc. [152]A. E. Dorigo, K. Morokuma, J. Am. Chem. Soc. 1989, 111, 4635 ± 1969, 91, 1852 ± 1854; J. Klein, R. M. Turckel, J. Am. Chem. Soc. 1969, 4643; A. E. Dorigo, K. Morokuma, J. Am. Chem. Soc. 1989, 111, 91, 6186 ± 6187; J. Klein, N. Aminadav, J. Chem. Soc. C 1970, 1380 ± 6524 ± 6536. 1385; Y. Yamamoto, H. Yatagai, K. Maruyama, J. Org. Chem. 1979, [153]E. Nakamura, Y. Miyachi, N. Koga, K. Morokuma, J. Am. Chem. Soc. 44, 1744 ± 1746; J. P. Marino, R. J. Linderman, J. Org. Chem. 1981, 46, 1992, 114, 6686 ± 6692; E. Nakamura, M. Nakamura, Y. Miyachi, N. 3696 ± 3702; I. Fleming, D. A. Perry, Tetrahedron 1981, 37, 4027 ± Koga, K. Morokuma, J. Am. Chem. Soc. 1993, 115, 99 ± 106. 4034. [154]C. W. Bauschlicher, Jr., S. R. Langhoff, H. Partridge, L. A. Barnes, J. [185]K. Nilsson, C. Ullenius, N. Krause, J. Am. Chem. Soc. 1996, 118, Chem. Phys. 1989, 91, 2399 ± 2411. 4194 ± 4195. [155]I. Antes, G. Frenking, Organometallics 1995, 14, 4263 ± 4268; [186]K. Nilsson, T. Andersson, C. Ullenius, J. Organomet. Chem. 1997, compare with: M. Böhme, G. Frenking, Chem. Phys. Lett. 1994, 546, 591 ± 595; K. Nilsson, T. Andersson, C. Ullenius, A. Gerold, N. 224, 195 ± 199. Krause, Chem. Eur. J. 1998, 4, 2051 ± 2508. [156]P. Belanzoni, M. Rosi, A. Sgamellotti, E. J. Baerends, C. Floriani, [187]J. Normant, G. Cahiez, J. Villieras, J. Organomet. Chem. 1975, 92, Chem. Phys. Lett. 1996, 257, 41 ± 48. C28-C30. [157]C. Sosa, J. Andzelm, B. C. Elkin, E. Wimmer, K. D. Dobbs, D. A. [188]Reviews: J. M. Klunder, G. H. Posner in Comprehensive Organic Dixon, J. Phys. Chem. 1992, 96, 6630 ± 6636. Synthesis, Vol. 3 &Eds.: B. M. Trost, I. Fleming), Pergamon, Oxford, [158]M. Böhme, G. Frenking, M. T. Reetz, Organometallics 1994, 13, 1991, pp. 207 ± 239. 4237 ± 4245. [189]R. W. Herr, D. M. Wieland, C. R. Johnson, J. Am. Chem. Soc. 1970, [159]H. O. House, W. L. Respess, G. M. Whitesides, J. Org. Chem. 1966, 92, 3813 ± 3814; C. R. Johnson, R. W. Herr, D. M. Wieland, J. Org. 31, 3128 ± 3141. Chem. 1973, 38, 4263 ± 4268; compare with: M. Aratani, L. V. [160]K. Yamamoto, H. Ogura, J.-i. Jukuta, H. Inoue, K. Hamada, Y. Dunkerton, T. Fukuyama, Y. Kishi, H. Kakoi, S. Sugiura, S. Inoue, Sugiyama, S. Yamada, J. Org. Chem. 1998, 63, 4449 ± 4458. J. Org. Chem. 1975, 40, 2009 ± 2011. [161]H. O. House, M. J. Umen, J. Am. Chem. Soc. 1972, 94, 5495 ± [190]M. J. Eis, B. Ganem, Tetrahedron Lett. 1985, 26, 1153 ± 1156. 5497. [191]a) C. R. Johnson, G. A. Dutra, J. Am. Chem. Soc. 1973, 95, 7783 ± [162]J. Munch-Petersen, C. Bretting, P. M. Jùrgensen, S. Refu, V. K. 7787; b) B. H. Lipshutz, R. S. Wilhelm, J. Am. Chem. Soc. 1982, 104, Andersen, Acta Chem. Scand. 1961, 15, 277 ± 292; see also: 4696 ± 4698. Y. Yamamoto, J. Yamada, T. Uyehara, J. Am. Chem. Soc. 1987, 109, [192]C.-y. Guo, M. L. Brownawell, J. San Filippo, Jr, J. Am. Chem. Soc. 5820 ± 5822. 1985, 107, 6028 ± 6030. [163]E. J. Corey, F. J. Hannon, N. W. Boaz, Tetrahedron 1989, 45, 545± [193]S. Komiya, T. A. Albright, R. Hoffmann, J. K. Kochi, J. Am. Chem. 555. Soc. 1976, 98, 7255 ± 7265. [164]A. S. Vellekoop, R. A. J. Smith, J. Am. Chem. Soc. 1994, 116, 2902 ± [194]S. Komiya, T. A. Albright, R. Hoffmann, J. K. Kochi, J. Am. Chem. 2913. Soc. 1976, 98, 8440 ± 8447. [165]S. R. Krauss, S. G. Smith, J. Am. Chem. Soc. 1981, 103, 141 ± 148. [195]A. Tamaki, J. K. Kochi, J. Organomet. Chem. 1973, 51, C39 ± C42. [166]H. O. House, K. A. J. Snoble, J. Org. Chem. 1976, 41, 3076 ± 3083. [196]J. Stein, J. P. Fackler, Jr., C. Paparizos, H.-W. Chen, J. Am. Chem. [167]C. P. Casey, M. C. Cesa, J. Am. Chem. Soc. 1979, 101, 4236 ± Soc. 1981, 103, 2192 ± 2198. 4244. [197]J. P. Collman, Acc. Chem. Res. 1968, 1, 136 ± 143. [168]S. H. Bertz, G. Dabbagh, J. M. Cook, V. Honkan, J. Org. Chem. 1984, [198]E. C. Ashby, R. N. DePriest, A. Tuncay, S. Srivastava, Tetrahedron 49, 1739 ± 1743. Lett. 1982, 23, 5251 ± 5254. [169]D. J. Hannah, R. A. J. Smith, Tetrahedron Lett. 1975, 187 ± 190; [199]E. C. Ashby, D. Coleman, J. Org. Chem. 1987, 52, 4554 ± 4565; S. H. R. A. J. Smith, D. J. Hannah, Tetrahedron 1979, 35, 1183 ± 1189. Bertz, G. Dabbagh, A. M. Mujsce, J. Am. Chem. Soc. 1991, 113, 631 ± [170]C. P. Casey, R. A. Boggs, Tetrahedron Lett. 1971, 27, 2455 ± 2458. 636. [171]G. M. Whitesides, P. E. Kendall, J. Org. Chem. 1972, 37, 3718 ± 3725; [200]H. L. Goering, C. C. Tseng, J. Org. Chem. 1983, 48, 3986 ± 3990; H. L. F. Näf, P. Degen, Helv. Chim. Acta 1971, 54, 1939 ± 1949. Goering, S. S. Kantner, J. Org. Chem. 1984, 49, 422 ± 426; H. L. [172]H. Yamataka, N. Fujimura, Y. Kawafuji, T. Hanafusa, J. Am. Chem. Goering, S. S. Kantner, E. P. Seitz, Jr., J. Org. Chem. 1985, 50, 5495 ± Soc. 1987, 109, 4305 ± 4308. 5499. [173]Y. Chounan, T. Ibuka, Y. Yamamoto, J. Chem. Soc. Chem. Commun. [201]a) M. A. Sevin, W. Chodkiewicz, P. Cadiot, Tetrahedron Lett. 1965, 1994, 2003 ± 2004. 1953 ± 1959; b) P. Rona, P. CrabbeÂ, J. Am. Chem. Soc. 1968, 90, 4733 ± [174]Y. Yamamoto, S. Nishii, T. Ibuka, J. Am. Chem. Soc. 1988, 110, 617 ± 4734; P. Rona, P. CrabbeÂ, J. Am. Chem. Soc. 1969, 91, 3289 ± 3292; 618. J. L. Luche, E. Barreiro, J. M. Dollat, P. CrabbeÂ, Tetrahedron Lett. [175]Y. Chounan, H. Horino, T. Ibuka, Y. Yamamoto, Bull. Chem. Soc. 1975, 4615 ± 4618; J.-M. Dollat, J.-L. Luche, P. CrabbeÂ, J. Chem. Soc. Jpn. 1997, 50, 1953 ± 1959. Chem. Commun. 1977, 761 ± 762; c) P. Vermeer, J. Meijer, L. [176]C. T. Wigal, J. R. Grunwell, J. Hershberger, J. Org. Chem. 1991, 56, Brandsma, Rec. Trav. Chim. Pays-Bas 1975, 94, 112 ± 114; A. 3759 ± 3761; S. J. Anderson, W. T. Hopkins, C. T. Wigal, J. Org. Chem. Alexakis, I. Marek, P. Mangeney, J. F. Normant, J. Am. Chem. Soc. 1992, 57, 4304 ± 4305. 1990, 112, 8042 ± 8047. [177]J. Berlan, J.-P. Battioni, K. Koosha, Bull. Chim. Soc. Fr. 1979, II183 ± [202]C. Gallina, P. G. Ciattini, J. Am. Chem. Soc. 1979, 101, 1035 ± 1036; 190. H. L. Goering, S. S. Kantner, C. C. Tseng, J. Org. Chem. 1983, 48, [178]G. Hallnemo, T. Olsson, C. Ullenius, J. Organomet. Chem. 1984, 265, 715 ± 721. C22 ± C24; G. Hallnemo, T. Olsson, C. Ullenius, J. Organomet. Chem. [203]J.-E. Bäckvall, M. Selle Ân, B. Grant, J. Am. Chem. Soc. 1990, 112, 1985, 282, 133 ± 144; B. Christenson, G. Hallnemo, C. Ullenius, Chem. 6615 ± 6621; E. S. M. Persson, J.-E. Bäckvall, Acta. Chem. Scand. Scr. 1987, 27, 511 ± 512; C. Ullenius, B. Christenson, Pure Appl. Chem. 1995, 49, 899 ± 906. 1988, 60, 57 ± 64. [204]E. Nakamura, K. Sekiya, M. Arai, S. Aoki, J. Am. Chem. Soc. 1989, [179]N. Krause, R. Wagner, A. Gerold, J. Am. Chem. Soc. 1994, 116, 381 ± 111, 3091 ± 3093. 382. [205]M. Nakamura, E. Nakamura, N. Koga, K. Morokuma, J. Chem. Soc. [180]S. Sharma, A. C. Oehlschlager, Tetrahedron 1991, 47, 1177 ± 1184; Faraday Trans. 1994, 90, 1789 ± 1798; K. Kubota, S. Mori, M. S. H. Bertz, R. A. J. Smith, J. Am. Chem. Soc. 1989, 111, 8276 ± 8277; Nakamura, E. Nakamura, J. Am. Chem. Soc. 1998, 120,13334±13341. B.-S. Lou, G. Li, F.-D. Lung, V. J. Hruby, J. Org. Chem. 1995, 60, [206]P. B. Armentrout, R. Georgiadis, Polyhedron 1988, 7, 1573 ± 1581. 5509 ± 5514. [207]D. E. Frantz, D. A. Singleton, J. P. Snyder, J. Am. Chem. Soc. 1997, [181]J. Canisius, A. Gerold, N. Krause, Angew. Chem. 1999, 111, 1727 ± 119, 3383 ± 3384. 1730; Angew. Chem. Int. Ed. 1999, 38, 1644 ± 1646. [208]B. E. Rossiter, N. M. Swingle, Chem. Rev. 1992, 92, 771 ± 806. [182]J. Klein, R. Levene, J. Chem. Soc. Perkin Trans. 2 1973, 1971 ± 1979. [209]Note that, in contrast to MeI, secondary iodides do not react in an

[183]N. Krause, Tetrahedron Lett. 1989, 30, 5219 ± 5222. SN2 manner. See Ref. [18].

3770 Angew. Chem. Int. Ed. 2000, 39, 3750 ± 3771 Organocopper Chemistry REVIEWS

[210]M. Nakamura, E. Nakamura, N. Koga, K. Morokuma, J. Am. Chem. Chem. Soc. 1998, 120, 5981 ± 5989; E. K. van den Beuken, B. L. Soc. 1993, 115, 11016 ± 11017. Feringa, Tetrahedron 1998, 12 985 ± 13 011. [211]S. Mori, B.-H. Kim, M. Nakamura, E. Nakamura, Chem. Lett. 1997, [215]See: H. Suzuki, H. Omori, Y. Moro-oka, Organometallics 1988, 7, 1079 ± 1080. 2579 ± 2581; H. Suzuki, H. Omori, D. H. Lee, Y. Yoshida, M. [212]E. Kaufmann, P. von R. Schleyer, K. N. Houk, Y.-D. Wu, J. Am. Fukushima, M. Tanaka, Y. Moro-oka, Organometallics 1994, 13, Chem. Soc. 1985, 107, 5560 ± 5562. 1129 ± 1146; M. Shibasaki, H. Sasai, T. Arai, Angew. Chem. 1997, 109, [213]M. P.Bernstein, D. B. Collum, J. Am. Chem. Soc. 1993, 115, 789 ± 790; 1290 ± 1310; Angew. Chem. Int. Ed. Engl. 1997, 36, 1236 ± 1256; see P. G. Willard, Q.-Y. Liu, J. Am. Chem. Soc. 1993, 115, 3380 ± 3381; also: T. Iida, N. Yamamoto, H. Sasai, M. Shibasaki, J. Am. Chem. Soc. F. E. Romesberg, D. B. Collum, J. Am. Chem. Soc. 1995, 117, 2166 ± 1997, 119, 4783 ± 4784; I. Ojima, M. Okabe, K. Kato, H. B. Kwon, I. T. 2178; K. W. Henderson, A. E. Dorigo, Q.-Y. Liu, P. G. Willard, HorvaÂth, J. Am. Chem. Soc. 1988, 110, 150 ± 157; H. Steinhagen, G. P. von R. Schleyer, P. R. Bernstein, J. Am. Chem. Soc. 1996, 118, Helmchen, Angew. Chem. 1996, 108, 2489 ± 2492; Angew. Chem. Int. 1339 ± 1347; compare with: E. L. CoitinÄ o, E. Ciuffarin, F. M. Floris, J. Ed. Engl. 1996, 35, 2339 ± 2342; S. J. Lippard, Science 1995, 268, 996 ± Tomasi, J. Phys. Chem. A 1998, 102, 8369 ± 8376. 997. [214]M. Schlosser, G. Jan, E. Byrne, J. Sicher, Helv. Chim. Acta 1973, 56, [216]E. Nakamura, A. Hirai, M. Nakamura, J. Am. Chem. Soc. 1998, 120, 1630 ± 1637; M. Schlosser, T. D. An, Angew. Chem. 1981, 93, 1114; 5844 ± 5845; A. Hirai, M. Nakamura, E. Nakamura, J. Am. Chem. Angew. Chem. Int. Ed. Engl. 1981, 20, 1039 ± 1041; M. G. Stanton, Soc. 1999, 121, 8665 ± 8666. C. B. Allen, R. M. Kissling, A. L. Lincoln, M. R. GagneÂ, J. Am.

Angew. Chem. Int. Ed. 2000, 39, 3750 ± 3771 3771